The Development of Joke and Irony Understanding: A Study With 3-to 6-Year-Old Children

Abstract

Previous research suggests that comprehending ironic utterances is a relatively late-developing skill, emerging around 5-6 years of age. This study investigated whether younger children might show an earlier understanding when ironic utterances are performed in familiar communicative situations, and investigated the relationships among irony comprehension, language, and theory of mind (ToM) abilities. A group of 100 children aged 3.0-6.5 years was presented with 4 types of puppet scenarios depicting different communicative interactions: control, joke, contingent irony and background irony stories. Results suggested that (a) even younger children easily understand jokes, and may sometimes understand ironies; (b) children's comprehension of irony continues to develop across early childhood; and (c) receptive vocabulary scores had simultaneous effects on irony comprehension and ToM performance. (PsycINFO Database Record (c) 2013 APA, all rights reserved).

Full text

The Development of Joke and Irony Understanding:
A Study With 3- to 6-Year-Old Children
Romina Angeleri and Gabriella Airenti
University of Turin
Previous research suggests that comprehending ironic utterances is a relatively late-developing skill,
emerging around 5– 6 years of age. This study investigated whether younger children might show an
earlier understanding when ironic utterances are performed in familiar communicative situations, and
investigated the relationships among irony comprehension, language, and theory of mind (ToM) abilities.
A group of 100 children aged 3.0 – 6.5 years was presented with 4 types of puppet scenarios depicting
different communicative interactions: control,joke,contingent irony and background irony stories.
Results suggested that (a) even younger children easily understand jokes, and may sometimes understand
ironies; (b) children’s comprehension of irony continues to develop across early childhood; and (c)
receptive vocabulary scores had simultaneous effects on irony comprehension and ToM performance.
Keywords: verbal irony, theory of mind, communication, developmental pragmatics
Supplemental materials: http://dx.doi.org/10.1037/cep0000011.supp
Verbal irony is a form of nonliteral communication and many
different definitions have been provided in the literature (Attardo,
2000;Clark & Gerrig, 1984;Kreuz & Glucksberg, 1989;Kumon-
Nakamura, Glucksberg, & Brown, 1995;Wilson & Sperber, 1992).
Developmental research has shown that the comprehension of
ironic utterances is a particularly complex task for children
(Capelli, Nakagawa, & Madden, 1990;Filippova & Astington,
2008;Winner, 1988). A number of studies have focused on the
different aspects connected with the development of children’s
understanding of this communicative act (for a review see
Creusere, 2000;Pexman & Glenwright, 2007). In particular, irony
comprehension has been compared with deceit and metaphor.
Some studies have investigated children’s difficulties in under-
standing intentional falsehood, which is typical of deceit, polite-
ness, and irony (Ackerman, 1981;Airenti & Angeleri, 2011;
Peterson, Peterson, & Seeto, 1983). It has been shown that children
as old as 13 years of age often fail to distinguish irony from
deception (Demorest, Meyer, Phelps, Gardner, & Winner, 1984;
Demorest, Silberstein, Gardner, & Winner, 1983). Data from
other studies indicate much earlier competence (e.g., Andrews,
Rosenblatt, Malkus, Gardner, & Winner, 1986), and the con-
clusion of most studies is that children’s comprehension of
irony starts between the age of 5 and 6 years (e.g., Dews &
Winner, 1997) and continues developing over time.
The particular difficulty in interpreting ironic utterances has
often been attributed to the complexity of the inferences involved
in comprehension, which are assumed to require a full-fledged
theory of mind (ToM). Several studies have shown that children’s
inability to grasp the meaning of ironic phrases may be connected
with their difficulties in inferring the speaker’s beliefs and inten-
tions (Sullivan, Winner, & Hopfield, 1995;Winner, Brownell,
Happé, Blum, & Pincus, 1998;Winner & Leekam, 1991). Accord-
ing to Winner (1988), in order to understand irony, the child has to
be able to detect incongruity or falsehood, infer motivation, and
attribute second-order beliefs to the speaker. Ackerman (1983)
suggested that in children’s comprehension of irony two indepen-
dent processes can be distinguished, that is, the detection of the
nonliteral form cued by contextual discrepancy and the process
of inferring the speaker’s intent cued by intonation. Hancock,
Dunham, and Purdy (2000) confirmed this dissociation and
attributed it to the fact that detection demands first-order rea-
soning about the speaker’s beliefs while inferring the speaker’s
intent requires inferences on the speaker’s beliefs about the
listener’s beliefs, that is, second-order reasoning.
The same developmental sequence has been confirmed by
Filippova and Astington (2008) in a study comparing children and
adults. These authors stress the fact that interpreting irony is
difficult for children and that this ability improves with age. In fact
even the 9-year-old participants in their study did not reach adult
skill levels. The progression was correlated with understanding of
mind and linguistic abilities. Filippova and Astington (2010) con-
firmed that the interpretation of complex mental states implied in
irony (i.e., speaker’s intention, motivation, and attitude) was dif-
ficult for children and developed beyond middle-school years.
However, they also found that for children interpreting the func-
tion of ironic utterances (i.e., how nice, mean, funny they are) was
This article was published Online First December 23, 2013.
Romina Angeleri and Gabriella Airenti, Center for Cognitive Science,
Department of Psychology, University of Turin, Turin, Italy.
This research was supported by PRIN Project 2008 (No. 2008N9KF5K)
from Ministero dell’Istruzione, dell’Universita
`e della Ricerca (MIUR).
We thank Valentina Berti, Miryam Donzelli, Andrea Dulicchio, and
Marika Tigani for their contribution in data collecting and coding, and Dr.
Marco Del Giudice for his valuable advice on the statistical analysis of the
results.
Correspondence concerning this article should be addressed to Romina
Angeleri, Department of Psychology, University of Turin, Via Po 14,
10123 Turin, Italy. E-mail: romina.angeleri@unito.it
Canadian Journal of Experimental Psychology / Revue canadienne de psychologie expérimentale © 2013 Canadian Psychological Association
2014, Vol. 68, No. 2, 133–146 1196-1961/14/$12.00 DOI: 10.1037/cep0000011
133

easier. This was viewed again in terms of theory of mind. One’s
own evaluation of the communicative effect of irony demands an
intuitive judgment on a situation while inferring a speaker’s com-
plex mental states demands metarepresentational reasoning.
In another study focused on the detection of verbal irony,
Nilsen, Glenwright, and Huyder (2011) presented children with
video recordings where two puppets interacted in different com-
municative contexts (literal criticism, ironic criticism, and literal
compliment), and then asked them questions about the listener’s
interpretations and beliefs. They found that 8- to 10-year-old
children were able to consider the listener’s knowledge state when
they had to interpret verbal irony. In particular, children of that age
performed similarly to adults when answering questions about the
listener’s beliefs, showing their ability to represent the listener’s
knowledge state and the listener’s interpretation of the speaker’s
statement, while younger children (6- to 7-year-olds) had difficul-
ties understanding the listener’s interpretation. Thus, in line with
previous research, they found that children’s second-order theory
of mind skills are related to their ability to interpret verbal irony,
allowing children to be more flexible in their interpretations.
Although several developmental studies have assessed chil-
dren’s understanding of irony during middle and late childhood,
there is still a need for data regarding irony comprehension in
younger children. To the best of our knowledge, there is only one
study that has shown an initial comprehension of irony in young
children (Loukusa & Leinonen, 2008). This study involved chil-
dren aged 3 to 9 years, and showed that even if the majority of
children could correctly interpret verbal irony only between 6 and
7 years of age, some 3- and 4-year-olds showed an emerging
ability to recognise the communicative intent in simple ironic
utterances. It is interesting that this study was conducted on Finn-
ish children, opening a window on cultural settings and languages
other than English.
All the results reported above are based on experimental studies
focused on the comprehension of ironic language, where children
had to understand and evaluate dialogues presented within stories,
cartoons, or puppet shows. Recently some works have tried to
address the issue of when and how children begin to produce irony,
examining various forms of irony during interactions in the family
context. Pexman, Zdrazilova, McConnachie, Deater-Deckard, and
Petrill (2009) studied the use of ironic gestures and utterances in
family triads of one parent and two children recorded while they
completed a cooperative dominos task. Their conclusion was that
irony production might begin at the same age as irony compre-
hension (the age of the youngest child to use verbal irony was 5.34
years). In a similar way, Recchia, Howe, Ross, and Alexander
(2010) examined the use of irony in conversations between parents
and their 4- and 6-year-old children at home. Their results showed
that even 4-year-old children occasionally used verbal irony, usu-
ally hyperbole, even if less frequently than their older siblings.
Thus, there is some evidence that children can begin to produce
ironic utterances at around 4 years of age.
Besides considering irony as a form of nonliteral communica-
tion, it is also possible to examine the discourse goals that are
accomplished by its use. In many cases, irony is used with humor-
ous intent (e.g., Long & Graesser, 1988;Roberts & Kreuz, 1994);
thus another possibility is to study the development of irony with
respect to the development of humor. The relationship between
irony and humor arises from studies conducted among adult sub-
jects, revealing that they consider ironic utterances funnier than
literal ones (e.g., Colston & Keller, 1998;Gibbs, 2000;Kreuz,
Long, & Church, 1991). As regards children, it has been suggested
that it is not before 7– 8 years of age that children start appreciating
the humorous aspect of irony (Dews et al., 1996;Harris & Pexman,
2003). Pexman, Glenwright, Krol, and James (2005) have tried to
find an explanation of this result. They suggested that children do
not appreciate the humor function of verbal irony in the same way
that adults do, as they are only beginning to share adults’ percep-
tions that ironic utterances (in particular, in case of ironic criti-
cism) are intended to be funnier than literal remarks. In their study,
children tended to identify themselves with the victims of criti-
cism, and had difficulties in dealing with conflicting representa-
tions of emotions and intentions.
From a theoretical point of view, in the pragmatics literature a
number of authors have tried to clarify the relation between irony
and humor (Attardo, 2002;Giora, 1995;Ritchie, 2005). Despite
the definitional problems, due to the partial overlapping of differ-
ent categories, irony is generally included in the broad category of
conversational humor (Attardo, 2002;Norrick, 2003). However,
little research has spanned both humor and irony (Attardo, 2002).
With respect to development, even if it seems useful to study
irony in connection with other forms of humor, this aim may be
difficult to achieve, given that while most studies on irony are
experimental and focused on comprehension, empirical work on
the development of humor in young children is mostly based on
naturalistic observations of interactions in family, daily care, and
nursery school (Bergen, 1989;Cameron, Kennedy, & Cameron,
2008;Groch, 1974;Hoicka & Akhtar, 2012;Loizou, 2005;Reddy,
2008). Therefore, the heterogeneity of both methods and commu-
nicative contexts makes comparing their results arduous.
However, we assume that considering irony on the background
of the wider phenomenon of humor can be suitable to understand
how children deal with different aspects of playful communication
since a young age.
Studies on young children have shown that simple forms of
humor appear very early in child development, reflecting how
important these emotional responses become in the interactions
with adults (Dunn, 1988). Laughter appears in infants at about 4
months (Sroufe & Piccard Wunsch, 1972) and at around 10 months
children both respond to humorous situations and produce humor-
ous acts themselves such as clowning and teasing (Reddy, 1991).
From a cognitive point of view, humor has been defined as the
discovery of some incongruity with respect to reality (McGhee,
1979;Shultz, 1976). From this perspective, the development of the
capacity of understanding and producing humor is related to chil-
dren’s ability to construct symbolic representations of reality,
which emerges at around 18 months of age, when children become
able to deal with fantasy and pretend play suggesting some abili-
ties to separate representations and realities. The relation between
children’s symbolic play and verbal humor has been supported by
a case study in which all the productions of verbal humor in a child
from 15 to 30 months of age have been systematically documented
as a function of conceptual complexity (Johnson & Mervis, 1997).
According to Shultz (1976), the real appreciation of humor de-
mands that children are able not only to represent incongruities,
but also to resolve them, and this ability emerges not before 6 years
of age.
134 ANGELERI AND AIRENTI

However, the necessity of representation and resolution of in-
congruity for humor appreciation remains controversial. Some
other authors have maintained that even very young children may
experience humor as far as they are able to detect incongruities in
reality, and to adopt a playful attitude. In this perspective it is an
expectancy violation in reality, like the mother failing to reappear
in a peek-a-boo game or someone slipping on a banana peel that is
perceived as funny (Pien & Rothbart, 1976). The playful attitude
makes possible for children to respond to incongruity with humor,
without confounding it with other emotions, as for example aston-
ishment or fear (Bariaud, 1989). The playful attitude of humor may
be in effect mingled with an aggressive component or be perceived
as aggressive (Veatch, 1998).
From an analysis of the literature we can conclude that irony
shares with other forms of humor some basic mechanisms:
a) They are all based on some form of incongruity or contrast
(Colston & O’Brien, 2000), as what the humorist does or says goes
against expectations in a funny way. With respect to irony, the
simplest situation is that of an utterance blatantly contradicting
reality as when someone says “Another gorgeous day!” while it is
raining heavily. There are several other kinds of irony forms but all
have in common the use of incongruity shared between the inter-
locutors to suggest a discrepancy between reality and expectation
(Gibbs, 1994).
b) Moreover, in humor as well as in irony there is often a latent
aggressive component. With respect to humor, teasing is the typ-
ical example (Keltner, Capps, Kring, Young, & Heerey, 2001), but
this is also the case of laughing at someone sliding on a banana
peel. Irony often implies criticism even if sometimes, but in fact
rather rarely, also compliments are expressed in an ironic form. In
fact there is a controversy about the function of irony with respect
to criticism. According to the tinge hypothesis, one function of
irony is precisely to decrease the negativity of criticisms (Dews,
Kaplan, & Winner, 1995). Other authors have found that on the
contrary irony enhances criticism (Colston, 1997;Toplak & Katz,
2000).
c) Finally, understanding the effects of humor and irony, re-
quires one to consider the expansion of the common ground
between the interlocutors (Clark, 1996). Adults and children share
the ability to detect “affordances for funniness” (Reddy, 2008).
This is what allows perceiving or constructing together humorous
situations and sharing amusement. As regards irony, the only way
to distinguish between ironic and serious utterances is by focusing
on shared attitudes and expectations (Airenti, Bara, & Colombetti,
1993). At the core of different theories of irony is the consideration
that irony is the reminder of (Kreuz & Glucksberg, 1989), the echo
of (Wilson & Sperber, 1992), the allusion to (Kumon-Nakamura et
al., 1995) something that interlocutors are supposed to share.
In conclusion, there are different definitions of irony, which
focus on different aspects of irony and its pragmatic function. We
consider that, instead of adopting one of them, we can operate
relying on a simplified definition that includes the main features
that the different models present in the literature have identified:
irony is a nonliteral utterance which demands to be understood
sharing a common ground, is focused on an unexpected incongru-
ity, and has a teasing aspect.
Starting from this simplified definition of irony we make the
general hypothesis that young children may understand the com-
municative intent of ironic utterances even if they are unable to
identify a utterance as ironic, that is, to explicitly explain their
understanding in terms of others’ mental states. Children could
directly access the communicative intent before being able to
attribute second-order beliefs. Actually, we try to depart from the
procedure utilized in most of the cited studies, in which children
were asked to evaluate utterances so as to judge the speaker’s
intentions and motivations, or to rank utterances on the basis of
how mean, or teasing, or funny they are. We move from the
assumption that young children could understand these aspects of
ironic utterances while being unable to verbally explain them. In
the Finnish study (Loukusa & Leinonen, 2008), children who
initially gave correct answers were not always able to subsequently
provide an explanation. This might clarify the discrepancy be-
tween the findings of experimental and observational studies; in
the latter, even young children appear to understand and produce
humorous and ironic sentences in the context of everyday interac-
tions. In this respect, Gibbs and O’Brien (1991) in their discussion
of the psychological understanding of irony made an interesting
point, arguing that people do not need to recognise irony to
comprehend what the speakers mean by their ironic utterances.
This conclusion was based on the finding that adults were able to
give correct paraphrase judgments for ironic expressions even if
they were not always conscious of their ironic character (Gibbs,
1986).
Children are precociously involved in humorous interactions
with parents. This develops children’s communicative abilities and
facilitates the construction of a common ground for comprehen-
sion. It is on this aspect that the present study is based.
Thus we proposed to distinguish between the ability to compre-
hend the communicative intent of ironic utterances—which, in our
perspective, relies on basic communicative abilities—and the abil-
ity to explicitly acknowledge the components and features of
irony—which demands second-order ToM abilities and only de-
velops later in childhood.
Taking the standpoint that irony is a form of humor involves the
question of what distinguishes one form of humor from another. A
theoretical analysis of all the different forms of humor goes be-
yond the scope of this paper. Actually, the boundaries between the
subcategories are fuzzy, no clear distinctions have been established
in the literature, and it is even unclear if providing definitions of
the different humorous phenomena is an attainable task (Attardo,
1994,2002;Norrick, 1993). Distinctions may involve the opposi-
tion between literal and nonliteral or the different degrees of
knowledge constituting the common ground requested for com-
prehension or how teasing they are. Moreover, different configu-
rations of these aspects can make the comprehension of some
forms more difficult than others.
Our standpoint that irony is a form of humor allowed us to
compare in a single study young children’s comprehension of
irony and joke. We have used the term joke as dictionaries define
it: something said or done for cause amusement or a trick played
for fun. Jokes result from the fact that what is said or done is
unexpected and then incongruous with respect to reality and are in
general mildly teasing. We considered that simple literal jokes
could be good instances of unsophisticated humor. As regards
irony, the traditional definition of Grice considered only the case
in which an utterance is taken to have a meaning opposite to its
literal content (Grice, 1975,1978). Most post-Gricean theories
adopt the point of view of Sperber and Wilson (1981), who argued
135
CHILDREN’S UNDERSTANDING OF JOKE AND IRONY

that there are several forms of irony where the intended meaning
is not simply the opposite of the literal one. With respect to
development it is possible that not all the forms of irony present
the same level of difficulty (Bosco & Bucciarelli, 2008). In our
study, we have distinguished two forms, contingent irony and
background irony, which we have supposed to be of different
difficulty with respect to comprehension due to the different access
to common ground they involve.
We assessed understanding presenting to 3.0- to 6.5-year-old
children a number of scenarios in which two puppets handled by
two experimenters were involved in four types of simple commu-
nicative interactions: (1) literal factual serious communicative acts
that can be conceived as control items (e.g., one puppet plays ball
and the other says: “I like playing ball too”), (2) joke, that is,
nonserious utterances (e.g., the puppets laughed at the experi-
menter making funny faces and one puppet said to the other “Look,
what funny faces!”), (3) contingent irony, the most simple case of
irony, based on what we call a form of manifest sharedness. The
irony in this case derives from the negation of something that is
directly perceived by the interlocutors, (e.g., one puppet fails to
score a basket and the other says: “Your shot was pretty good!”),
and finally (4) background irony, a different case of irony, based
on what we can call a form of previously acquired sharedness (e.g.,
one puppet breaks a plate and the other says: “Your mommy will
be happy!”), where irony is based on something that the interloc-
utors are supposed to share but which is not directly perceived or
mentioned (in this case, the fact that the mother is not happy when
her child breaks something).
It is important to note that children were presented with simple
interactions depicting familiar situations (e.g., puppets playing
with sand, playing with a basket, and so on), and we asked them
open questions with the purpose of creating a relatively naturalistic
communicative situation, which offered the opportunity to provide
expansive and comprehensive answers. Our objective was to have
children providing their interpretation of the communicative acts
instead of orienting them to give judgments using predefined
aspects of irony. Moreover, we proposed to put forward the hu-
morous component of irony placing side-by-side ironic utterances
and joking situations. In particular we focused on common ground,
that is, shared knowledge that constitutes the basis for comprehen-
sion (Clark, 1996). We considered that in situations where the
common ground is easy to deal with, even young children would
be able to understand jokes, and probably also some forms of
irony.
Our expectations were the following:
1) All children— even the younger ones—would be able to
understand the humorous intent underlying joke situations.
2) At least some children in the younger groups would succeed
in the irony comprehension task. We suggest that children
are able to participate in communicative interactions in
which irony is in place not because they understand the exact
nature of the ironic utterances (i.e., the use of nonliteral
language, the ironic intentions of the speaker, and so on), but
because they participate in a wide range of communicative
interactions, becoming more and more familiar with the use
of communicative practices and activities. Thus, if the chil-
dren are presented with familiar interactions, their perfor-
mances in the experimental situations should be more similar
to the ones observed in everyday life.
3) Joke and irony understanding would show different stages of
development. We predicted that joke is the easiest form,
followed in turn by contingent irony, and finally by back-
ground irony. The basis of this prediction was the assump-
tion that jokes make a very early part of young children
interactions with adults and are based on incongruities im-
mediately perceived while irony is constructed on the basis
of more elaborated knowledge. In turn, the common ground
shared by the participants in the contingent situation is more
direct than in the background situation, which appears more
demanding.
4) Finally, in order to investigate the critical factors involved in
irony understanding, we explored its relationship with ToM
abilities and child receptive vocabulary as a measure of
language development, two aspects that the literature has
shown as strongly related (e.g., Lewis & Osborne, 1990;
Milligan, Astington, & Dack, 2007;O’Neill, 2005).
Method
Participants
A total of 100 Italian children (50 girls, 50 boys) were tested.
All children were recruited through kindergarten centers and ele-
mentary schools from the greater Turin area. The mean age for
children was 4.77 years (SD !1.14 years). The global sample was
divided into the following four age groups: 25 3.0- to 3.5-year-olds
(M!3.34 years; SD !2.1 month), 25 4.0- to 4.5-year-olds (M!
4.25 years; SD !1.8 months), 25 5.0- to 5.5-year-olds (M!5.19
years; SD !3.88 months) and 25 6.0- to 6.5-year-olds (M!6.32
years; SD !2.88 months). For the sake of simplicity, we will refer
to the four age groups as 3-year-olds, 4-year-olds, 5-year-olds and
6-year-olds. Each age group included an equal number of boys and
girls. Criteria for inclusion were that participants had no history of
speech and language difficulties, were Italian native speakers, and
had no known significant medical or neurological condition. Par-
ents were informed about the research details and provided in-
formed consent for their children to participate in our study.
In order to check the consistency of language ability in each age
group, children were administered the Peabody Picture Vocabulary
Test–Revised (PPVT-R; Dunn & Dunn, 1981; Italian adaptation:
Stella, Pizzioli, & Tressoldi, 2000). The mean scores for each
group are reported with other participants’ details in Table 1. All
children were within the normal range. Moreover, to ensure ho-
mogeneity of linguistic ability within each age group, Grubb’s test
was used in the statistical evaluation of the data to identify outliers.
For the PPVT-R, no outliers were detected at the 95% significance
level within each age group, consequently we did not exclude any
participants from the sample. As it will be clear in the Result
section, the scores obtained using the PPVT-R were further anal-
ysed to examine the contribution of verbal skills in irony under-
standing.
Socioeconomic status (SES) was measured by family composi-
tion, parental education level and occupation, and was obtained
using the Two-Factor Index of Social Position (ISP) developed by
136 ANGELERI AND AIRENTI

Hollingshead (1975), combining information about parents’ jobs
and educational attainment. Parents were free to decide whether
they would complete the SES questionnaires; nine families did not
return the questionnaire. According to the parent reports of the
remaining 91 families, the majority of children were from the
middle social class (47.2%), but other classes were also repre-
sented (lower: 9.7%; lower-middle: 25%; upper-middle: 13.9%;
and upper: 4.2%). Only 4.2% of children lived with one single
parents, while 95.8% of children lived with two married or cohab-
iting parents.
Materials
Irony task. Sixteen puppet show scenarios were depicted; we
created scenarios that were likely to be familiar to the 3.0- to
6.5-year-old children (e.g., playing with sand, eating cakes, going
for a bike ride). Some examples of the puppet scenarios are given
in the Appendix (available online as supplemental material). In
each scenario, two puppets were involved in a communicative
interaction; the speaker puppet did not perform any action but
produced an utterance in response to the other puppet’s behaviour.
The utterance might have been a simple and common comment
(control situation), a joke, or an ironic utterance; the utterances
were equivalent in terms of length and syntactical difficulties. In
more detail, there were four different types of scenarios (four
scenarios for each type):
1) Control stories, where the speaker puppet produced a literal,
factual, and serious comment spoken with matter-of-fact
tone in response to the other puppet’s behaviour (contextual
cues present);
2) Joke, where the speaker puppet produced an utterance spo-
ken with joking tone to evoke laughter or amusement (con-
textual cue present);
3) Contingent irony, where the speaker puppet produced an
ironic utterance spoken with joking tone that implied the
negation of something directly perceived by the interlocutors
(contextual cue present);
4) Background irony, where the speaker puppet produced an
ironic utterance spoken with joking tone that could be com-
prehended only referring to shared knowledge; for example,
normally children don’t like spinach, any mom tries to stop
sibling squabbles, and so on (contextual cue absent). The
proposed situations are expected to refer to contexts in which
children are involved from very early on and therefore
familiar and shared among all young children.
Theory of mind tasks. Three ToM tasks were used in the
present study. These tasks examined the ability to understand first-
and second-order false belief. First-order false belief tasks required
a child to make an inference concerning a false belief about a state
of the world; to examine this ability we used the following two
tasks: the Smarties task (Perner, Leekam, & Wimmer, 1987), and
the Sally-Ann task (Wimmer & Perner, 1983).
In the Smarties task, children were shown a familiar container
(in this case, a Pringles box, due to the fact that Smarties boxes are
no longer familiar to Italian children) and were asked to indicate
what it held. Then the unusual contents were revealed (i.e., a
pencil) and the children were asked what a person who had not
seen the pencil would have said was inside the container.
For the Sally-Ann task, we followed the experimental procedure
suggested by Baron-Cohen, Leslie, and Frith (1985). We presented
children with two dolls, Sally and Ann, and first we asked children
the naming question (“Who is this doll? And this one?”) in order
to check that the children knew which doll was which. Then Sally
placed a ball into her basket, and left the scene; the ball was
transferred by Ann and hidden in her box. When Sally returned,
the experimenter asked the critical belief question: “Where will
Sally look for her ball?” If the children replied/pointed to the
previous location of the ball, then they correctly responded to the
belief question by appreciating Sally’s present false belief. If they
replied/pointed to the ball’s current location, then they failed the
belief question by not taking into account the doll’s belief. These
conclusions were warranted if two control questions were an-
swered correctly: “Where is the marble really?” (reality question);
“Where was the marble in the beginning?” (memory question).
Second-order stories investigate the ability to understand a false
belief about another character’s belief; to examine this ability we
used the Ice-cream van story task (Baron-Cohen, 1989). The story
was read to the children who were shown a paperboard model
scenario presenting the action sequences involved in the story.
John and Mary are together in the park. Along comes the ice-cream
man. John would like to buy an ice cream but has no money with
him. The ice-cream man tells him to go home and get his money.
In the meantime he will be staying in the park. When John comes
home to get the money, the ice-cream man moves to the church.
Later John meets the ice-cream man in front of the church, but
Mary does not know about that because she came back home
before. Children were then asked: “Where does Mary think that
John has gone to buy an ice-cream?”
Table 1
Sample Description: Demographic Details and Peabody Scores of Participants
Age group No. of children
Sex Age (years) Peabody score
Female Male M SD M SD
3-yr-olds 25 12 13 3.34 .17 38.20 9.65
4-yr-olds 25 13 12 4.25 .15 55.56 13.71
5-yr-olds 25 13 12 5.19 .32 80.88 19.16
6-yr-olds 25 12 13 6.32 .24 95.48 18.93
Total 100 50 50 Mean 4.77 .22 67.53 15.36
137
CHILDREN’S UNDERSTANDING OF JOKE AND IRONY

Procedure
Children were tested individually in a separate quiet room
outside of their classroom. The PPVT-R and all theory of mind
tests were presented first. Then, two puppets were introduced to
the children. The various objects involved in the experimental
scenarios were all placed in a small bag; during the experimental
sessions, they were taken from the bag and put on the table in front
of the children. After every puppet show, each child was asked:
“Why did the puppet say that?” Throughout the interviews, the
experimenter used some prompts to elucidate children’s responses
to the questions and/or to obtain explanations of the responses; the
prompt questions were the same for each child, but they were used
only when the child was briefly distracted or the responses were
ambiguous, and they did not necessarily indicate a failure in the
comprehension task.
In order to further clarify the experimental protocol used in the
present study, we report some examples of items:
(1) Control Story
Puppet A is drawing; puppet B says: “So nice you’re drawing!”
Examiner: “Why did puppet B say that?”
If the child’s answer is not clear, ambiguous, or merely descriptive,
the examiner gives a prompt. For example, if the child replies “He’s
drawing,” the examiner asks: “And why did the puppet say that?”
(2) Joke
Puppet A makes funny faces; the two puppets are laughing. Puppet B
says: “Look! What funny faces!”
Examiner: “Why did Puppet B say that?”
If the child’s answer is not clear, ambiguous, or merely descriptive,
the examiner gives a prompt. For example:
If the child replies “Because he makes funny faces,” the examiner
asks: “And why did the puppet say that?”
(3) Contingent Irony
Puppet A fails to score a basket; puppet B says: “Well done!”
Examiner: “Why did puppet B say that?”
If the child’s answer is not clear, ambiguous, or merely descriptive,
the examiner gives a prompt. For example:
If the child replies: “Because they are playing,” the examiner asks:
“And why did the puppet say that?”
If the child repeats: “Well done” the examiner asks: Was it well done?
If the child replies “No,” the examiner asks: “And why did the puppet
say that?”
(4) Background Irony
Puppet A broke a plate; puppet B says: “Your mom will be very
happy!”
Examiner: “Why did puppet B say that?”
If the child’s answer is not clear, ambiguous, or merely descriptive,
the examiner gives a prompt. For example:
If the child replies “Because he has broken a plate” the examiner asks:
“And why did the puppet say that?”
If the child replies “The mommy will be happy,” the examiner asks:
“Will mommy be happy?” If the child replies “No,” the examiner
asks: “And why did the puppet say that?”
Narratives for the puppet scenarios were told by two experi-
menters who had been instructed to use the appropriate tone
according to the situations. For prompts the experimenters were
instructed to use a neutral tone in order to exclude any cue toward
a preferred response, and to stop after two prompts. The puppet
scenarios were presented in random order.
The order of the 16 scenarios was randomized across children.
The experimental sessions lasted approximately 40 minutes, per-
mitting pauses if needed. All sessions were audiotaped and chil-
dren’s responses verbatim were transcribed.
Some more examples of experimental questions and children’s
responses are given in the Appendix (available online as supple-
mental material).
Scoring Procedure
Irony task. After every puppet scenario, there was a question
(“Why did the puppet say that?”) involving children’s understand-
ing of the communicative intent. Only in the case that children
made ambiguous replies, the examiner posed a set of other ques-
tions aimed at clarifying children’s understanding. Children’s re-
plies obtained 1 point if they understood the joke/ironic commu-
nicative intent; otherwise, they obtained a score of 0. Responses
were considered correct if the children identified the correct com-
municative intention as resulting from the explicit reference to the
puppet’s utterance, in order to avoid the possibility that children
were generically referring to the proposed situation.
We present here some examples of scoring.
(1) Control Story
Puppet A is drawing; puppet B says: “So nice you’re drawing!”
Example of child’s answer: “Because he likes drawing” (score !1)
Comment: In the answer the child recognises the seriousness of the
utterance.
Example of child’s answer: “I don’t know why” (score !0)
(2) Joke
Puppet A makes funny faces; the two puppets are laughing. Puppet B
says: “Look! What funny faces!”
Example of child’s answer: “He said that because it is so fun”
(score !1)
Comment: In the answer the child recognises that the puppet’s utter-
ance is supposed to evoke amusement.
Example of child’s answer: “Because they are laughing” (score !0)
Comment: The child does not refer to the utterance but to the
situation.
(3) Contingent Irony
Puppet a fails to score a basket; Puppet B says: “Well done!”
138 ANGELERI AND AIRENTI

Example of child’s answer: “Because he wanted to make fun of him”
(score !1)
Comment: The child recognises that the puppet’s answer is nonliteral
and nonserious.
Example of child’s answer: “Because he was good at playing”
(score !0)
Comment: The child takes the puppet’s answer literally.
(4) Background Irony
Puppet A broke a plate; puppet B says: “Your mom will be very
happy!”
Example of child’s answer: “The mom will be mad!” (score !1)
Comment: The child recognises that the puppet’s answer is nonliteral.
Example of child’s answer: “The mom will be happy” (Examiner:
“Will mommy be happy?”) Child: Yes (score !0)
Comment: The child takes the puppet’s answer literally.
Two independent raters coded children’s responses. The Co-
hen’s kvalue was .90, indicating almost perfect agreement (Landis
& Koch, 1977).
The Appendix (available online as supplemental material) dis-
plays some other examples of children’s answers, and scoring.
Theory of mind tasks. The theory of mind tasks were coded
on the basis of the coding procedures normally used in the liter-
ature. In more detail, in the Smarties task children’s answers were
coded as correct if they replied “Pringles!,” showing an appreci-
ation of the mental state of the person involved in the task (Perner,
Leekam, & Wimmer, 1987). In the Sally-Ann task an error in ToM
is revealed at the belief question if the child answers the wrong
location of the ball (i.e., the box), since Sally is unaware of Ann’s
covert action (Wimmer & Perner, 1983). To pass the task and
obtain 1 point at the belief question, the children had to pass all the
questions posed (naming, reality, and memory questions; Baron-
Cohen et al., 1985). In the Ice-cream van story task children
obtained 1 point if they passed test question and if they correctly
answered the control questions, following the procedure reported
in Wimmer and Perner (1983).
To summarise, for each task a single score was given for
children’s responses (0 –1). Also in this case, the responses of all
participants on every theory of mind task were coded by two
independent judges; interrater agreement was almost perfect (Co-
hen’s k!.94).
Analysis and Results
Comparison Between Irony Scenarios
First, children’s scores across irony scenarios and age groups
were compared to determine whether there was a difference among
the tasks, and whether the ability to understand different types of
irony was characterised by an age-related improvement. Table 2
displays the mean proportions of correct responses for the four
types of scenarios (control, joke, contingent irony, and background
irony stories) in the four age groups and in the whole sample.
To test the hypothesis that different tasks would show increasing
difficulty levels, we performed a repeated measures ANOVA on
the whole sample (see the bottom line of Table 2). A linear contrast
was fitted to the data under the hypothesis that task difficulty
would be ordered as follows: control stories "jokes "contingent
irony "background irony. The analysis showed a significant
effect of task difficulty in the predicted direction (F
(1, 99)
!
255.37, p".001, #
2
!.72).
The effect of age on children’s performance was investigated
with a between-subjects ANOVA with four levels corresponding
to the four age groups, and average performance across tasks as the
dependent variable (rightmost column of Table 2). A linear con-
trast was fitted to the data under the hypothesis that performance
would increase linearly with age. Again, age showed a significant
effect on performance in the predicted direction (F
(1, 98)
!31.02,
Table 2
Mean (SD) of Correct Responses to the Different Types of Scenarios (Scoring Range 0 –1)
Age group Control stories Joke Contingent irony
Background
irony Average score
3-yr-olds .97 (.08) .85 (.24) .57 (.33) .40 (.26) .70 (.16)
4-yr-olds 1.00 (.00) .87 (.21) .71 (.30) .52 (.25) .77 (.12)
5-yr-olds 1.00 (.00) 1.00 (.00) .87 (.22) .60 (.24) .87 (.09)
6-yr-olds .95 (.12) .97 (.09) .92 (.14) .63 (.23) .86 (.09)
Whole sample .98 (.08) .92 (.17) .77 (.29) .54 (.26) .80 (.14)
Figure 1. Children’s performance on the control, joke, contingent, and
background irony stories by age group.
139
CHILDREN’S UNDERSTANDING OF JOKE AND IRONY

p".001, #
2
!.24), indicating that children’s performance sig-
nificantly increased with increasing age (see Figure 1).
Correlations Between Irony, Language, and
ToM Scores
First, in order to investigate the relation between children’s ToM
development and their understanding of irony, correlational anal-
yses were performed. Children’s scores on the various ToM mea-
sures (Sally & Ann, Smarties’ task, and Ice-cream story) were
correlated with understanding of irony tasks (control, joke, con-
tingent irony, and background irony). Table 3 presents the descrip-
tive statistics for each ToM task in the sample. Table 4 displays
correlations among ToM tasks and irony comprehension.
Statistically significant correlations were found among the ToM
tasks (.31 "r".35; p".01), as well as among the irony tasks,
namely joke, contingent irony, and background irony (.23 "r"
.46; p".05). No significant correlations were observed between
control stories and ToM tasks ($.14 "r"$.03; p%.05), and
with irony stories ($.12 "r".05; p%.05). For this reason, in
the following analysis we collapsed the three ToM tasks into a
single ToM score, and the three pragmatic tasks (joke, contingent
irony and background irony) into a single humor comprehension
score, while the control stories were excluded. While the three
pragmatic tasks are conceptually distinct, collapsing them into a
single score provides a more statistically robust measure of overall
pragmatic ability (Cronbach’s alpha !.73), as well as more robust
estimates of path coefficients. Indeed, running separate analyses
for each task (humor, contingent irony, and background irony)
yielded virtually identical patterns of results, supporting the deci-
sion to employ the summary score in the main analysis.
In order to examine what factors contribute to success in irony
comprehension, the global humor comprehension score was then
correlated with children’s performance on PPVT-R (language
test), the global ToM score, chronological age, and sex. Table 5
shows correlations among these variables.
The humor score was significantly correlated with the ToM
score (r!.37, p".0001), with the PPVT-R score (r!.56, p"
.0001), and with children’s age (r!.54, p".0001); no significant
correlations were found with the sex of participants. While this
pattern of correlations could imply a causal effects of both ToM
and language abilities on humor comprehension, the interpretation
of these findings is complicated by the correlation between ToM
and language, and— even more importantly— between all the per-
formance variables and chronological age. On the one hand, indi-
vidual differences in language abilities might determine a spurious
correlation between ToM and humor comprehension. On the other
hand, the whole pattern of correlations might be explained by age
differences between children. In order to disentangle these possi-
bilities and ascertain the specific effects of ToM and language on
humor comprehension, we fitted a series of path analysis models to
the data and employed formal model selection techniques to
choose the most appropriate among them.
Path Analysis: Effects of ToM, Language Abilities,
and Age on Humor Comprehension
To begin with, a set of four alternative path-analytic models
(Models A$D in Figure 2) was constructed based on theoretical
considerations. In these models, various combinations of age,
ToM, and language ability (PPVT-R) predicted humor compre-
hension, either directly or indirectly. In all models, age had direct
causal effects on language, ToM, and humor comprehension. In
Model A, both language and ToM had direct effects on humor
comprehension; moreover, language had a direct effect on ToM
(and thus an additional indirect effect on humor comprehension).
Model B was the same as Model A without the effect of language
on ToM. In Model C, only ToM had a direct effect on humor
comprehension, while language had only an indirect effect through
ToM. Finally, in Model D, language ability affected both humor
comprehension and ToM, and ToM had no independent causal
effect on humor comprehension.
Models were fitted to the covariance matrix using maximum-
likelihood estimation, and compared using the small-sample ver-
sion of Akaike’s AIC (AIC
C
; see Burnham &Anderson, 2002).
Models were fit with R 2.8.0 (R Development Core Team, 2008)
with the sem package version 3.0 (Fox, 2008). Fit indices are
shown in Figure 2. Model D achieved the best combination of fit
and parsimony, as shown by the lowest value of AIC
C
, and was
therefore selected as the best model in the set. In order to test the
Table 4
Correlations Between Irony and ToM Scores (Global Sample)
Variables 1 2 3 4 5 6 7
1. Sally & Ann —
2. Smarties’ task .35
!!
—
3. Ice-cream story .31
!!
.34
!!
—
4. Control stories $.07 $.03 $.14 —
5. Joke .14 .23
!
.12 $.12 —
6. Contingent irony .34
!!
.32
!!
.21
!
$.01 .24
!
—
7. Background irony .15 .27
!!
.11 .05 .23
!
.46
!!
—
!
p".05.
!!
p".01.
Table 5
Correlations of Irony Score, ToM Score, Peabody Score, Age,
and Sex (Global Sample)
Variables 1 2 3 4 5
1. Irony score —
2. ToM score .37
!!
—
3. PPVT-R .56
!!
.61
!!
—
4. Age .54
!!
.62
!!
.81
!!
—
5. Sex .10 .05 .01 .01 —
!
p".05.
!!
p".01.
Table 3
Mean (SD) of Correct Responses to the Theory of Mind Tasks
(Scoring Range 0 –1)
Age group
Sally &
Ann
Smarties’
task
Ice-cream
story Total
3-yr-olds .08 (.28) .20 (.41) .00 (.00) .09 .15
4-yr-olds .40 (.50) .25 (.44) .04 (.20) .25 .29
5-yr-olds .64 (.49) .63 (.49) .28 (.46) .51 .33
6-yr-olds .84 (.37) .72 (.46) .36 (.49) .64 .30
Mean .49 (.50) .45 (.50) .17 (.38) .37 (.35)
140 ANGELERI AND AIRENTI

possible effects of children’s sex, direct effects of sex on ToM,
language, and humor comprehension were added to Model D,
leading to Model E (see Figure 2). Model E failed to outperform
Model D, and sex was consequently dropped from the analysis.
The standardized parameters of Model D are shown in Figure 3.
As expected, age significantly predicted both language ability
and ToM skills, while the direct effect of age on humor compre-
hension was not significant. Language had simultaneous effects on
humor comprehension and ToM performance; however, ToM had
no independent effect on humor comprehension. In other words, in
the selected model the correlation between ToM and humor un-
derstanding (see Table 5) was entirely spurious. Specifically, this
correlation was accounted for by (a) the shared direct effects of
language ability on ToM and humor, and (b) the shared indirect
effects of chronological age on language and ToM.
Discussion
The current study was designed to investigate the emergence of
humor and irony comprehension in young children. Our goal was
to examine the possibility of an early competence and a progres-
sive increment already during the preschool period; moreover, we
tried to clarify the role played by theory of mind and language
skills in children’s acquisition of these communicative acts.
A number of experimental studies have shown that irony is a
communicative phenomenon that is difficult to grasp. The obser-
Figure 2. Potential models among age, language skills, ToM ability, and irony comprehension.
Figure 3. Standardized parameters of model D (see Figure 2). Variances,
disturbances, and measurement errors are omitted for clarity of presenta-
tion.
141
CHILDREN’S UNDERSTANDING OF JOKE AND IRONY

vation of real life situations seems to contradict this fact: is it true
that when a mother tells her child coming back dirty from the
playroom “Here you are, ready for a visit to your grandmother!”
the child does not understand the real meaning intended by the
mother? A mother who utters such an utterance implies shared
knowledge about a familiar situation for the child, being washed
and changed clothes before a visit to grandmother. The general
purpose of our study was to investigate whether the difficulty that
many children displayed in experimental situations might be as-
cribed to the fact that children were asked to judge communicative
acts and to make the comprehension of their ironic nature explicit.
Actually, it has been shown that even adults may well understand
the communicative intent of irony without taking notice of the
specific ironic form used (Gibbs, 1986). We can then consider that
this be true also for children. Understanding what is meant by
ironic utterances should be assessed without asking children ex-
plicit judgments, which by definition imply second-order theory of
mind abilities.
For that reason, we designed a task in which children had to
show their comprehension of the communicative intent of ironic
utterances—that is, what the speaker meant— by answering open-
ended questions related to natural communicative contexts. We
assumed that adopting such a flexible approach children would be
able to demonstrate their emergent irony understanding. Moreover,
as irony can be considered as a form of humor, we associated irony
comprehension to simple joke situations.
Children were required to deal with communicative situations
that are rather common in parent/children and peers interactions.
These situations covered different forms of humor, including jokes
and two forms of irony, and some everyday serious interactions
used as control items. We considered that the knowledge necessary
for the comprehension of these situations could be assumed with a
reasonable probability. We expected that under these conditions
the task would be accessible also to young children, and that at
least some of them would have been able to understand the
communicative intent even in the case of ironic utterances.
In the present study, the performance of children in all age
groups in jokes did not differ significantly from the performance in
control situations. As regards the comprehension of irony we
found an increasing trend of performance with children’s age; as
expected the immediate access to shared knowledge on which
irony is based makes the understanding of contingent irony easier
than the understanding of background irony. From our point of
view, the most interesting result regards irony understanding in the
younger groups of children. In fact, the 3- and 4-year-olds also
showed a good performance in the contingent irony tasks (.64 and
.75, respectively) and a reasonable comprehension of background
irony (.45 and .55, respectively). Thus our results showed that even
3- and 4-year-old children might comprehend the actual intent of
an ironic communicative act. An interesting element is that our
data show an important individual variability with regards partic-
ularly to the younger groups. It is in these groups that we find
children who gave correct answers to all the items and others who
never showed comprehension. Different studies on the develop-
ment of humor have shown a high degree of variability within
groups of children (Brodzinsky & Righmyer, 1980;Masten, 1989;
St. James & Tager-Flusberg, 1994;Varga, 2000). Moreover, this is
consistent with the Finnish study (Loukusa & Leinonen, 2008) that
noted the same effect in children aged 3 to 6, an effect that the
authors related with a different use of irony in home environment.
This is also consistent with our hypothesis that comprehension of
irony is not dependent on the ability of making inferences on the
other’s mind, but on the familiarity with these specific forms of
communicative interactions. While ToM abilities show a certain
degree of individual variability, they also increase strongly with
age. On the contrary, as Recchia and colleagues (2010) have
found, there is a great deal of variability across families in their use
of ironic language. As these authors suggest, to gain new insight
on this correlation future research should combine children’s ob-
servation in family context and experimental assessment.
Our results show that it is useful posing the distinction between
the comprehension of the communicative intent of an ironic utter-
ance and understanding of the ironic form. The fact that there are
two different steps in irony comprehension has been proposed in
the literature (Ackerman, 1983;Hancock et al., 2000) attributing
these two steps to different levels of ToM abilities. Our point of
view is more radical. Children are involved since very young in
communicative interactions, which imply shared knowledge
(Airenti, 2010). It is on the background of this familiar knowledge
that children may have direct access to the speaker’s communica-
tive intent without elaborating the specific literal phrasing. In fact,
people in general, even adults in everyday communicative inter-
actions, just limit to comprehend the “real meaning” of a commu-
nicative act, which often is not the literal meaning. Adults are more
proficient with respect to children because in case of failure or
uncertainty they may resort to more complex inferences while this
is not the case for young children. Our approach is compatible with
those studies that have shown that young children are able to grasp
the communicative intentions of their interlocutors while having
difficulties taking account of the specific linguistic form that has
been used to convey them (Olson & Hildyard, 1981;Robinson,
Goelman & Olson, 1983). It has been suggested that the capacity
of distinguishing literal meaning from communicative intent is a
general ability that develops between first and second grades (Beal
& Flavell, 1984).
In the literature about irony this issue has been summarised as
the contrast between one-stage and two-stage theories (Attardo,
2000). On this matter a debate has opposed Gibbs (1984), who has
contended that the comprehension of the real meaning is immedi-
ate and does not pass through the failure of the literal meaning, and
other authors assuming that there are reasons to think that speakers
have access also to the literal meaning (Dews & Winner, 1995;
Giora, 1997). Recently this debate has been resumed by Gibbs and
Colston (2012), who argue against a principled distinction between
literal and nonliteral language and maintain that figurative lan-
guage does not represent any cognitive deviation. Actually, this is
a general question regarding the link between language and com-
munication. A great part of human communication is not literal
and it is a fact that adults deal with it rather easily. Surely, there are
cases where the doubt arises if an utterance has to be taken for
instance as a literal praise or as an ironic blame. What can make the
doubt arise is a lack of shared knowledge. Think of someone saying
“Paul is so brilliant!” The interlocutor may not be sure about the
knowledge and judgments possibly shared with the speaker, and so
may be unable to decide if the utterance has to be taken as ironic
or not. If the interlocutor, on the contrary, shares with the speaker
a negative judgment about Paul, the ironic interpretation is imme-
diately activated as the literal one would be in the case they shared
142 ANGELERI AND AIRENTI

a positive judgment. Thus, questioning about the kind of utterance
produced by the speaker, literal or nonliteral, is linked to a failure
of interpreting it on the background of shared knowledge (Airenti
et al., 1993). In conclusion, comprehension does not require fully
developed ToM abilities, while producing judgments on commu-
nicative acts is a ToM task. In the comprehension of the commu-
nicative intent children, like adults, rely on other factors such as
their familiarity with the situation and the accessibility of the
knowledge that constitutes the background of the communicative
act. This last aspect is particularly relevant for irony comprehen-
sion, as shared knowledge is the only element leading toward the
nonliteral interpretation. In addition, children are able to coordi-
nate multiple cues in order to interpret ironic utterance, including
the literal statement itself, the speaker’s personality, and the type
of statement, from very early (Climie & Pexman, 2008). The
variability of results especially among the younger children can
be considered as supporting this point of view. We found the same
variability in a study of ours presently in preparation in which we have
used parent reports to give account of 2- to 7-year-old children’s
production of jokes, contingent irony, and background irony. Also, in
this case variability and a notable precocity in dealing with irony
shown by some of the children were associated.
The theoretical perspective and the experimental design that we
have adopted— choosing to include in the same study both jokes
and irony tasks—takes a stand on the relation between irony and
humor more in general. We believe that there is a strong link
between these two phenomena. In fact, in the developmental
literature such a link has been formulated mainly in terms of irony
components, ascertaining the capacity to single out the humorous
component of irony. Dews and colleagues (1996) and Harris and
Pexman (2003) showed that children appreciate rather late the
humorous aspect of irony, while Filippova and Astington (2010)
found that even adults did not find irony particularly funny under
the same experimental conditions. Our results showed that if the
problem is posed in terms of the comprehension of the communi-
cative intent, even young children could comprehend that the
incongruity of an ironic utterance is intended for fun. This does not
entail that they find ironic utterances funny. As we mentioned in
the introduction, irony as humor in general has always a latent
component of aggressiveness. Thus the perception of fun in humor
depends on which side one is in real situations (the humorist or the
victim) or with which character one identifies in stories, as it has
been shown by Pexman and colleagues (2005). Here we enter a
very interesting field which goes beyond the simple comprehen-
sion of communicative acts and which surely deserves further
work.
A fundamental point we intended to tackle in our study con-
cerned the relations among theory of mind abilities, language
skills, and irony understanding, an issue that has received much
attention in the current literature (Milligan et al., 2007). From the
present study, both language and ToM skills positively correlated
with irony understanding; while this picture could be consistent
with a causal effect of both cognitive abilities on irony understand-
ing, its interpretation is complicated by the fact that language and
ToM skills are correlated to each other, and that both are related to
chronological age. Our analyses suggested that the correlation
between irony understanding and ToM was spurious, accounted by
the shared effects of language ability on ToM and irony, and by the
shared indirect effects of children’s age on language and ToM.
This finding indicates that simple correlations between ToM, lan-
guage ability, and irony understanding should be interpreted with
considerable caution, since at least some of the effects may be
partially or entirely spurious. In order to avoid misleading inter-
pretations of research findings, explicit causal models should be
constructed and tested; in particular, the independent effects of
ToM and language should be assessed after controlling for their
statistical overlap and for the indirect effects of chronological age.
Although the ToM abilities in our study seem to play a minor
role, these findings are in line with those by Filippova and
Astington (2008) with respect to language development, and
with those of several other studies in showing the important role
of language in the acquisition of theory of mind (see for
instance, Dunn & Brophy, 2005;Dunn, Brown, Slomkowski,
Tesla, & Youngblade, 1991;Nelson, 2005;Nelson et al., 2003;
Ruffman, Slade, & Crowe, 2002). Considering that the most
relevant aspect in irony understanding is the access to shared
knowledge, we can interpret this result considering that linguis-
tic proficiency allows a more active role in communicative
interactions thus facilitating the acquisition of knowledge nec-
essary to comprehension. In conclusion, we can argue that when
actively involved in communication, even young children may
implicitly understand the communicative intentions of their
interlocutors, before they acquire complete ToM skills, as mea-
sured by the well-known ToM tasks.
In the present study we introduced a distinction between irony
based on directly perceived information and irony based on shared
knowledge. While our findings support the usefulness of this
dichotomy, we intend it as a heuristic device rather than an
exhaustive taxonomy of ironic acts and we expect that future
research in this area will reveal other useful distinctions between
related forms of humorous communication.
In our experimental procedure, children were not asked to
provide explicit judgments on the nature of the communicative
acts they observed. Clearly, a limitation of this approach is that we
cannot know exactly how children interpreted those acts. It could
be that more specific questions would elicit more explicit infor-
mation about the nature of children’s interpretations. However, our
procedure was chosen to be as close as possible to real-life situa-
tions, in which children are rarely if ever asked to precisely
explicate their understanding of a humorous event or interaction.
The novel contribution of the present study is that we were able
to document young children’s ability to deal with irony, by dis-
tinguishing between the ability to comprehend the speaker’s mean-
ing in communicative contexts from the ability to give explicit
judgments on the nature of his or her utterances. By focusing on
the former, we can begin to explain why young children appear
way more proficient in real life than they do in most experimental
settings.
Résumé
Les recherches antérieures suggèrent que la compréhension
d’énoncés ironiques est une capacité qui se développe relativement
tard, vers l’âge de 5 ou 6 ans. La présente étude cherchait a
`
déterminer si les enfants plus jeunes pouvaient témoigner d’une
compréhension d’énoncés ironiques exprimés dans le cadre de
situations de communication familières. En outre, elle cherchait a
`
déterminer les relations possibles entre la compréhension de
143
CHILDREN’S UNDERSTANDING OF JOKE AND IRONY

l’ironie, le langage et les processus cognitifs de la théorie de
l’esprit. On a présenté a
`un groupe de 100 enfants, âgés de 3,0 a
`6,5
ans, 4 types de scénarios mettant en scène des marionnettes pour
illustrer différentes interactions communicationnelles : situation
contrôle, blague, ironie contextuelle et ironie basée sur des con-
naissances antérieures. Les résultats obtenus suggèrent a) que les
enfants plus jeunes comprennent aisément les blagues et parfois
l’ironie; b) la compréhension de l’ironie se développe au cours de
la petite enfance; c) les scores pour les mots compris avaient des
effets a
`la fois sur la compréhension de l’ironie et le rendement sur
le plan de la théorie de l’esprit.
Mots-clés : ironie verbale, théorie de l’esprit, communication,
pragmatique développementale.
References
Ackerman, B. (1981). Young children’s understanding of a speaker’s
intentional use of a false utterance. Developmental Psychology, 17,
472– 480. doi:10.1037/0012-1649.17.4.472
Ackerman, B. (1983). Form and function in children’s understanding of
ironic utterances. Journal of Experimental Child Psychology, 35, 487–
508. doi:10.1016/0022-0965(83)90023-1
Airenti, G. (2010). Is a naturalistic theory of communication possible?
Cognitive Systems Research, 11, 165–180. doi:10.1016/j.cogsys.2009.03
.002
Airenti, G., & Angeleri, R. (2011). Dealing with sincerity: Pathways of
communication in young children. British Journal of Developmental
Psychology, 29, 765–782. doi:10.1348/2044-835X.002007
Airenti, G., Bara, B. G., & Colombetti, M. (1993). Failures, exploitations
and deceits in communication. Journal of Pragmatics, 20, 303–326.
doi:10.1016/0378-2166(93)90055-T
Andrews, J., Rosenblatt, E., Malkus, U., Gardner, H., & Winner, E. (1986).
Children’s abilities to distinguish metaphoric and ironic utterances from
mistakes and lies. Communication & Cognition, 19, 281–297.
Attardo, S. (1994). Linguistic theories of humor. Berlin, Germany: Mouton
de Gruyter.
Attardo, S. (2000). Irony as relevant inappropriateness. Journal of Prag-
matics, 32, 793– 826. doi:10.1016/S0378-2166(99)00070-3
Attardo, S. (2002). Humor and irony in interaction: From mode adoption to
failure of detection. In L. Anolli, R. Ciceri, & G. Riva (Eds.), Say not to
say: New perspectives on miscommunication (pp. 159 –179). Amster-
dam, Netherlands: IOS Press.
Bariaud, F. (1989). Age differences in children’s humor. Journal of Chil-
dren in Contemporary Society, 20, 15– 45. doi:10.1300/J274v20n01_03
Baron-Cohen, S. (1989). The autistic child’s theory of mind: A case of
specific developmental delay. Journal of Child Psychology and Psychi-
atry, 30, 285–297. doi:10.1111/j.1469-7610.1989.tb00241.x
Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child
have a “theory of mind”? Cognition, 21, 37– 46. doi:10.1016/0010-
0277(85)90022-8
Beal, C. R., & Flavell, J. H. (1984). Development of the ability to
distinguish communicative intention and literal message meaning. Child
Development, 55, 920 –928. doi:10.2307/1130143
Bergen, D. (1989). Characteristics of young children’s expression of hu-
mour in home settings as observed by parents. International Journal of
Educology, 3, 124 –135.
Bosco, F. M., & Bucciarelli, M. (2008). Simple and complex deceits and
ironies. Journal of Pragmatics, 40, 583– 603. doi:10.1016/j.pragma
.2007.05.004
Brodzinsky, D. M., & Righmyer, J. (1980). Individual differences in
children’s humour development. In P. McGhee & A. Chapman (Eds.),
Children’s humour (pp. 181–212). Chichester, UK: Wiley.
Burnham, K. P., & Anderson, D. R. (2002). Model selection and multi-
model inference (2nd ed.). New York, NY: Springer-Verlag.
Cameron, E. L., Kennedy, K. M., & Cameron, C. A. (2008). “Let me show
you a trick!: A toddler’s use of humor to explore, interpret, and negotiate
her familial environment during a Day in the Life.Journal of Research
in Childhood Education, 23, 5–18. doi:10.1080/02568540809594642
Capelli, C. A., Nakagawa, N., & Madden, C. M. (1990). How children
understand sarcasm: The role of context and intonation. Child Develop-
ment, 61, 1824 –1841. doi:10.2307/1130840
Clark, H. H. (1996). Using language. Cambridge, UK: Cambridge Uni-
versity Press. doi:10.1017/CBO9780511620539
Clark, H. H., & Gerrig, R. J. (1984). On the pretense theory of irony.
Journal of Experimental Psychology: General, 113, 121–126. doi:
10.1037/0096-3445.113.1.121
Climie, E. A., & Pexman, P. M. (2008). Eye gaze provides a window on
children’s understanding of verbal irony. Journal of Cognition and
Development, 9, 257–285. doi:10.1080/15248370802247939
Colston, H. L. (1997). Salting a wound or sugaring a pill: The pragmatic
functions of ironic criticism. Discourse Processes, 23, 25– 45. doi:
10.1080/01638539709544980
Colston, H. L., & Keller, S. B. (1998). You’ll never believe this: Irony and
hyperbole in expressing surprise. Journal of Psycholinguistic Research,
27, 499 –513. doi:10.1023/A:1023229304509
Colston, H. L., & O’Brien, J. (2000). Contrast and pragmatics in figurative
language: Anything understatement can do, irony can do better. Journal
of Pragmatics, 32, 1557–1583. doi:10.1016/S0378-2166(99)00110-1
Creusere, M. A. (2000). A developmental test of theoretical perspectives
on the understanding of verbal irony: Children’s recognition of allusion
and pragmatic insincerity. Metaphor and Symbol, 15, 29 – 45. doi:
10.1207/S15327868MS151&2_3
Demorest, A., Meyer, C., Phelps, E., Gardner, H., & Winner, E. (1984).
Words speak louder than actions: Understanding deliberately false re-
marks. Child Development, 55, 1527–1534. doi:10.2307/1130022
Demorest, A., Silberstein, L., Gardner, H., & Winner, E. (1983). Telling it
as it isn’t: Children understanding of figurative language. British Journal
of Developmental Psychology, 1, 121–134. doi:10.1111/j.2044-835X
.1983.tb00550.x
Dews, S., Kaplan, J., & Winner, E. (1995). Why not say it directly? The
social functions of irony. Discourse Processes, 19, 347–367. doi:
10.1080/01638539509544922
Dews, S., & Winner, E. (1995). Muting the meaning: A social function of
irony. Metaphor and Symbolic Activity, 10, 3–19. doi:10.1207/
s15327868ms1001_2
Dews, S., & Winner, E. (1997). Attributing meaning to deliberately false
utterances: The case of irony. In C. Mandell & A. McCabe (Eds.), The
problem of meaning: Behavioral and cognitive perspectives (pp. 377–
414). Amsterdam, Netherlands: North-Holland/Elsevier Science. doi:
10.1016/S0166-4115(97)80142-2
Dews, S., Winner, E., Kaplan, J., Rosenblatt, E., Hunt, M., Lim, K., . . .
Smarsh, B. (1996). Children’s understanding of the meaning and func-
tions of verbal irony. Child Development, 67, 3071–3085. doi:10.2307/
1131767
Dunn, J. (1988). The beginnings of social understanding. Cambridge, MA:
Harvard University Press.
Dunn, J., & Brophy, M. (2005). Communication, relationships, and indi-
vidual differences in children’s understanding of mind. In J. W. Asting-
ton & J. Baird (Eds.), Why language matters to theory of mind (pp.
50 – 69). Oxford, UK: Oxford University Press. doi:10.1093/acprof:oso/
9780195159912.003.0003
Dunn, J., Brown, J., Slomkowski, C., Tesla, C., & Youngblade, L. (1991).
Young children’s understanding of other people’s feelings and beliefs:
Individual differences and their antecedents. Child Development, 62,
1352–1366. doi:10.2307/1130811
144 ANGELERI AND AIRENTI

Dunn, L. M., & Dunn, L. M. (1981). Peabody Picture Vocabulary Test-
Revised. Circle Pines, MN: American Guidance Service.
Filippova, E., & Astington, J. W. (2008). Further development in social
reasoning revealed in discourse irony understanding. Child Develop-
ment, 79, 126 –138. doi:10.1111/j.1467-8624.2007.01115.x
Filippova, E., & Astington, J. W. (2010). Children’s understanding of
social-cognitive and social-communicative aspects of discourse irony.
Child Development, 81, 913–928. doi:10.1111/j.1467-8624.2010
.01442.x
Fox, J. (2008). The SEM Package (Version 0.9 –13) [Software]. Retrieved
from http://cran.r-project.org/
Gibbs, R. W. Jr. (1984). Literal meaning and psychological theory. Cog-
nitive Science, 8, 275–304. doi:10.1207/s15516709cog0803_4
Gibbs, R. W. Jr. (1986). On the psycholinguistics of sarcasm. Journal of
Experimental Psychology: General, 115, 3–15. doi:10.1037/0096-3445
.115.1.3
Gibbs, R. W. Jr. (1994). The poetics of mind. Cambridge, UK: Cambridge
University Press.
Gibbs, R. W. Jr. (2000). Irony in talk among friends. Metaphor and
Symbol, 15, 5–27. doi:10.1080/10926488.2000.9678862
Gibbs, R. W. Jr., & Colston, H. L. (2012). Interpreting figurative meaning.
Cambridge, UK: Cambridge University Press. doi:10.1017/
CBO9781139168779
Gibbs, R. W. Jr., & O’Brien, J. (1991). Psychological aspects of irony
understanding. Journal of Pragmatics, 16, 523–530. doi:10.1016/0378-
2166(91)90101-3
Giora, R. (1995). On irony and negation. Discourse Processes, 19, 239 –
264. doi:10.1080/01638539509544916
Giora, R. (1997). Understanding figurative and literal language: The
graded salience hypothesis. Cognitive Linguistics, 8, 183–206.
Grice, H. P. (1975). Logic and conversation. In P. Cole & J. L. Morgan
(Eds.), Syntax and semantics 3: Speech acts (pp. 41–58). New York,
NY: Academic Press.
Grice, H. P. (1978). Further notes on logic and conversation. In P. Cole
(Ed.), Syntax and semantics 9: Pragmatics (pp. 113–128). New York,
NY: Academic Press.
Groch, A. S. (1974). Joking and appreciation of humor in nursery school
children. Child Development, 45, 1098 –1102. doi:10.2307/1128101
Hancock, J. T., Dunham, P. J., & Purdy, K. (2000). Children’s com-
prehension of critical and complimentary forms of verbal irony.
Journal of Cognition and Development, 1, 227–248. doi:10.1207/
S15327647JCD010204
Harris, M., & Pexman, P. M. (2003). Children’s perceptions of the social
functions of verbal irony. Discourse Processes, 36, 147–165. doi:
10.1207/S15326950DP3603_1
Hoicka, E., & Akhtar, N. (2012). Early humour production. British Journal
of Developmental Psychology, 30, 586 – 603. doi:10.1111/j.2044-835X
.2011.02075.x
Hollingshead, A. B. (1975). Four factors index of social status. New
Haven, CT: Yale University Press.
Johnson, K. E., & Mervis, C. B. (1997). First steps in the emergence of
verbal humor: A case study. Infant Behavior and Development, 20,
187–196. doi:10.1016/S0163-6383(97)90021-7
Keltner, D., Capps, L., Kring, A. N., Young, R. C., & Heerey, E. A. (2001).
Just teasing: A conceptual analysis and empirical review. Psychological
Bulletin, 127, 229 –248. doi:10.1037/0033-2909.127.2.229
Kreuz, R. J., & Glucksberg, S. (1989). How to be sarcastic: The echoic
reminder theory of verbal irony. Journal of Experimental Psychology:
General, 118, 374 –386. doi:10.1037/0096-3445.118.4.374
Kreuz, R. J., Long, D. L., & Church, M. B. (1991). On being ironic:
Pragmatic and mnemonic implications. Metaphor and Symbolic Activity,
6, 149 –162. doi:10.1207/s15327868ms0603_1
Kumon-Nakamura, S., Glucksberg, S., & Brown, M. (1995). How about
another piece of pie: The allusional pretence theory of discourse irony.
Journal of Experimental Psychology: General, 124, 3–21. doi:10.1037/
0096-3445.124.1.3
Landis, J. R., & Koch, G. G. (1977). The measurement of observer
agreement for categorical data. Biometrics, 33, 159 –174. doi:10.2307/
2529310
Lewis, C., & Osborne, A. (1990). Three-year-olds’ problems with false
belief: Conceptual deficit or linguistic artifact? Child Development, 61,
1514 –1519. doi:10.2307/1130760
Loizou, E. (2005). Infant humor: The theory of the absurd and the em-
powerment theory. International Journal of Early Years Education, 13,
43–53. doi:10.1080/09669760500048329
Long, D. L., & Graesser, A. C. (1988). Wit and humor in discourse
processing. Discourse Processes, 11, 35– 60. doi:10.1080/
01638538809544690
Loukusa, S., & Leinonen, E. (2008). Development of comprehension of
ironic utterances in 3- to 9-year-old Finnish-speaking children. Psychol-
ogy of Language and Communication, 12, 55– 69. doi:10.2478/v10057-
008-0003-0
Masten, A. S. (1989). Humor appreciation in children: Individual differ-
ences and response sets. Humor, 2, 365–384. doi:10.1515/humr.1989.2
.4.365
McGhee, P. (1979). Humor: Its origin and development. San Francisco,
CA: Freeman and Company.
Milligan, K., Astington, J. W., & Dack, L. A. (2007). Language and theory
of mind: Meta-analysis of the relation between language ability and false
belief understanding. Child Development, 78, 622– 646. doi:10.1111/j
.1467-8624.2007.01018.x
Nelson, K. (2005). Language pathways into the community of mind. In
J. W. Astington & J. Baird (Eds.), Why language matters to theory of
mind (pp. 26 – 49). Oxford, UK: Oxford University Press. doi:10.1093/
acprof:oso/9780195159912.003.0002
Nelson, K., Skwerer, D. P., Goldman, S., Henseler, S., Presler, N., &
Walkenfeld, F. F. (2003). Entering a community of minds: An experi-
ential approach to “theory of mind”. Human Development, 46, 24 – 46.
doi:10.1159/000067779
Nilsen, E. S., Glenwright, M., & Huyder, V. (2011). Children and adults
understand that verbal irony interpretation depends on listener knowl-
edge. Journal of Cognition and Development, 12, 374 – 409. doi:
10.1080/15248372.2010.544693
Norrick, N. R. (1993). Conversational joking. Bloomington, IN: Indiana
University Press.
Norrick, N. R. (2003). Issues in conversational joking. Journal of Prag-
matics, 35, 1333–1359. doi:10.1016/S0378-2166(02)00180-7
Olson, D. R., & Hildyard, A. (1981). Assent and compliance in children’s
language. In W. P. Dickson (Ed.), Children’s oral communication skills
(pp. 313–335). New York, NY: Academic Press.
O’Neill, D. K. (2005). Talking about “new” information: The given/new
distinction in children’s developing theory of mind. In J. W. Astington
& J. A. Baird (Eds.), Why language matters for theory of mind (pp.
84 –105). Oxford, UK: Oxford University Press.
Perner, J., Leekam, S. R., & Wimmer, H. (1987). Three-year-olds’ diffi-
culty with false belief: The case for a conceptual deficit. British Journal
of Developmental Psychology, 5, 125–137. doi:10.1111/j.2044-835X
.1987.tb01048.x
Peterson, C. C., Peterson, J. L., & Seeto, D. (1983). Developmental
changes in ideas about lying. Child Development, 54, 1529 –1535. doi:
10.2307/1129816
Pexman, P. M., & Glenwright, M. (2007). How do typically developing
children grasp the meaning of verbal irony? Journal of Neurolinguistics,
20, 178 –196. doi:10.1016/j.jneuroling.2006.06.001
Pexman, P. M., Glenwright, M., Krol, A., & James, T. (2005). An acquired
taste: Children’s perceptions of humor and teasing in verbal irony.
Discourse Processes, 40, 259 –288. doi:10.1207/s15326950dp4003_5
145
CHILDREN’S UNDERSTANDING OF JOKE AND IRONY

Pexman, P. M., Zdrazilova, L., McConnachie, D., Deater-Deckard, K., &
Petrill, S. A. (2009). “That was smooth, mom”: Children’s production of
verbal and gestural irony. Metaphor and Symbol, 24, 237–248. doi:
10.1080/10926480903310286
Pien, D., & Rothbart, M. K. (1976). Incongruity and resolution in chil-
dren’s humor: A re-examination. Child Development, 47, 966 –971.
doi:10.2307/1128432
R Development Core Team. (2008). R: A language and environment for
statistical computing (Software). Vienna, Austria: R Foundation for
Statistical Computing. Retrieved from http://www.R-project.org
Recchia, H. E., Howe, N., Ross, H. S., & Alexander, S. (2010). Children’s
understanding and production of verbal irony in family conversations.
British Journal of Developmental Psychology, 28, 255–274. doi:
10.1348/026151008X401903
Reddy, V. (1991). Playing with others’ expectations: Teasing and mucking
about in the first year. In A. Whiten (Ed.), Natural theories of mind:
Evolution, development and simulation of everyday mindreading (pp.
143–158). Oxford, UK: Basil Blackwell.
Reddy, V. (2008). How infants know minds. Cambridge, MA: Harvard
University Press.
Ritchie, D. (2005). Frame-shifting in humor and irony. Metaphor and
Symbol, 20, 275–294. doi:10.1207/s15327868ms2004_3
Roberts, R. M., & Kreuz, R. J. (1994). Why do people use figurative
language? Psychological Science, 5, 159 –163. doi:10.1111/j.1467-9280
.1994.tb00653.x
Robinson, E. J., Goelman, H., & Olson, D. R. (1983). Children’s under-
standing of the relation between expressions (what was said) and inten-
tions (what was meant). British Journal of Developmental Psychology,
1, 75– 86. doi:10.1111/j.2044-835X.1983.tb00545.x
Ruffman, T., Slade, L., & Crowe, E. (2002). The relation between chil-
dren’s and mothers’ mental state language and theory-of-mind under-
standing. Child Development, 73, 734 –751. doi:10.1111/1467-8624
.00435
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, NY: Wiley.
Sperber, D., & Wilson, D. (1981). Irony and the use-mention distinction. In
P. Cole (Ed.), Radical pragmatics (pp. 295–318). New York, NY:
Academic Press.
Sroufe, L. A., & Piccard Wunsch, J. (1972). The development of laughter
in the first year of life. Child Development, 43, 1326 –1344. doi:10.2307/
1127519
Stella, G., Pizzioli, C., & Tressoldi, P. (2000). Peabody–Test di vocabo-
lario recettivo. Torino, Italy: Omega Edizioni.
St. James, P. J., & Tager-Flusberg, H. (1994). An observational study of
humor in autism and Down Syndrome. Journal of Autism and Develop-
mental Disorders, 24, 603– 617. doi:10.1007/BF02172141
Sullivan, K., Winner, E., & Hopfield, N. (1995). How children tell a lie
from a joke: The role of second-order mental state attributions. British
Journal of Developmental Psychology, 13, 191–204. doi:10.1111/j
.2044-835X.1995.tb00673.x
Toplak, M., & Katz, A. N. (2000). On the uses of sarcastic irony. Journal
of Pragmatics, 32, 1467–1488. doi:10.1016/S0378-2166(99)00101-0
Varga, D. (2000). Hyperbole and humor in children’s language play.
Journal of Research in Childhood Education, 14, 142–151. doi:10.1080/
02568540009594759
Veatch, T. C. (1998). A theory of humor. Humor, 11, 161–215. doi:
10.1515/humr.1998.11.2.161
Wilson, D., & Sperber, D. (1992). On verbal irony. Lingua, 87, 53–76.
doi:10.1016/0024-3841(92)90025-E
Wimmer, H., & Perner, J. (1983). Beliefs about beliefs: Representation and
constraining function of wrong beliefs in young children’s understand-
ing of deception. Cognition, 13, 103–128. doi:10.1016/0010-
0277(83)90004-5
Winner, E. (1988). The point of words. Cambridge, MA: Harvard Univer-
sity Press.
Winner, E., Brownell, H., Happé, F., Blum, A., & Pincus, D. (1998).
Distinguishing lies from jokes: Theory of mind deficits and discourse
interpretation in right hemisphere brain-damaged patients. Brain and
Language, 62, 89 –106. doi:10.1006/brln.1997.1889
Winner, E., & Leekam, S. (1991). Distinguishing irony from deception:
Understanding the speaker’s second-order intention. British Journal of
Developmental Psychology, 9, 257–270. doi:10.1111/j.2044-835X.1991
.tb00875.x
Received May 15, 2013
Accepted September 28, 2013 !
146 ANGELERI AND AIRENTI