26 July 2021
AperTO - Archivio Istituzionale Open Access dell'Università di Torino
Communicative abilities in children: An assessment through different phenomena and expressive means / Francesca
Marina Bosco; Romina Angeleri; Livia Colle; Katiuscia Sacco; Bruno G. Bara. In: JOURNAL OF CHILD LANGUAGE.
-ISSN 0305-0009. - STAMPA. - 40(2013), pp. 741-778.
Original Citation:
Communicative abilities in children: An assessment through different phenomena and expressive means
Published version:
DOI:10.1017/S0305000913000081
Terms of use:
Open Access
(Article begins on next page)
Anyone can freely access the full text of works made available as "Open Access". Works made available under a
Creative Commons license can be used according to the terms and conditions of said license. Use of all other works
requires consent of the right holder (author or publisher) if not exempted from copyright protection by the applicable law.
Availability:
This is the author's manuscript
inSAperTO
UNIVERSITÀ
DEGLI STUDI
DI TORINO
This is the author's final version of the contribution published as:
Francesca Marina Bosco; Romina Angeleri; Livia Colle; Katiuscia Sacco;
Bruno G. Bara. Communicative abilities in children: An assessment through
different phenomena and expressive means. JOURNAL OF CHILD
LANGUAGE. 40 pp: 741-778.
DOI: 10.1017/S0305000913000081
The publisher's version is available at:
http://www.journals.cambridge.org/abstract_S0305000913000081
When citing, please refer to the published version.
Link to this full text:
http://hdl.handle.net/2318/76016
This full text was downloaded from iris - AperTO: https://iris.unito.it/
iris - AperTO
Com m unicative abilities in children: An assessm ent
through different phenom ena and expressive m eans*
F R A N C E S C A M . B O S C O
Center fo r Cognitive Science, Department o f Psychology, University o f Turin, and Neuroscience Institute o f Turin
R O M I N A A N G E L E R I
Center fo r Cognitive Science, Department o f Psychology, U niversity o f Turin
L I V I A C O L L E
Center fo r Cognitive Science, Department o f Psychology, University o f Turin, and Neuroscience Institute o f Turin
K A T I U S C I A S A C C O
Center fo r Cognitive Science, Department o f Psychology, University o f Turin, Neuroscience Institute o f Turin, and C C S f M R I Neuroradiology at K oelliker
H ospital, Turin
A N D
B R U N O G . B A R A
Center fo r Cognitive Science, Department o f Psychology, University o f Turin, and Neuroscience Institute o f Turin
A B S T R A C T
Previous studies on children’s pragm atic abilities have tended to focus on ju st one pragm atic phenomenon and one expressive m eans at a time, m ainly concentrating on com prehension, and overlooking the p ro duction side. We assessed both com prehension and production in relation to several pragm atic phenomena (simple and com plex standard com m unication acts, irony, and deceit) and several expressive means (linguistic, extralinguistic, paralinguistic). O ur study involved 390 Italian-speaking children divided into three age grou ps: 5 ;0 - 5 ;6 , 6 ; 6 - 7 ;o, and 8; 0 -8 ; 6. C hild ren’ s perform ance on all tasks im proved
with their age. W ithin each age group, children responded more accurately to tasks involving standard com m unication than to those involving deceit and irony, across all expressive means and for both com prehension and production. W ithin each pragm atic phenomenon, children responded more accurately to sim ple acts than to com plex ones, regardless o f age group and expressive means, i.e., linguistic or extralinguistic. O verall results fit well with the Cognitive Pragm atics theory (Bara, 2010).
I N T R O D U C T I O N
T h e goal o f the present study is tw ofold: we w ish to provide a com pre hensive picture o f children’s pragm atic abilities, which have generally been researched separately in the literature using different theoretical fram eworks and experim ental protocols, and to outline a theoretical explanation for the developm ental pathw ays observed.
Com m unicative/pragm atic ability refers to the use o f an expressive means to convey m eaning in specific contexts. T h is hum an ability allows us to realize different pragm atic phenomena, for exam ple indirect speech acts, i.e., acts through which the speaker com municates to the partner more than what he actually says (Searle, 19 75), deceit, i.e., intentional attempts to m anipulate the listener’ s mental state in order to induce her/him to believe something untrue about the reality (Perner, 19 9 1), and irony, i.e., com m unicative acts expressing the opposite o f what is m eant by the speaker (G rice, 1989). T h e m ain expressive means used to produce pragm atic phenomena are: (a) linguistic, i.e., the system atic use o f written and spoken w ord s; (b) extralinguistic, for exam ple the use o f gestures and facial expression s; and (c) paralinguistic, for exam ple the tone and the rhythm of the voice. A ll these aspects can be expressed and tested, from an em pirical perspective, in both com prehension and production. T o the best o f our knowledge no overall assessment o f how children’s pragm atic abilities are displayed across all these aspects exists in the current literature.
A lthough there are num erous excellent experim ental studies investigating children’ s com m unicative ability, as a rule these have focused on a single pragm atic phenomenon, such as, for exam ple, indirect speech acts (e.g., Bernicot & Legros, 1987), deceit (e.g., Sodian, 19 9 1; London & Nunez, 2002; T alw ar & L ee, 2008), irony (e.g.. D ew s et al., 19 9 6 ; C reusere, 2000; Pexm an & G lenw right, 2007), and sarcasm (G len w right & Pexm an, 2010). M oreover, even when they have considered m ore than one pragm atic phenomenon at the same time (e.g.. W inner & Leekm an, 19 9 1; G io ra & F ein , 19 9 9 ; A irenti & A ngeleri, 2 0 1 1), they focused on a single expressive m eans at a time, usually the linguistic one, with few exceptions (e.g., Bucciarelli, Colle & Bara, 2003). In the clinical context, a variety o f
exhaustive clinical tools have been developed, such as standardized tests (Adam s, Cooke, C rutchley, Hesketh & R eeves, 20 0 1), checklists and profiles (D ew art & Sum m ers, 19 9 7 ; Bishop, 1998), and tasks assessing referential com munication (L lo yd , Peers & Foster, 20 01). H ow ever, these tools typically focus on language, neglecting extralinguistic and paralinguistic com m unicative m odalities.
T h u s, existing studies tend to focus on ju st one pragm atic phenomenon at a time, and/or consider only one expressive means - usually the linguistic one - lim iting the investigation to com prehension, and overlooking the production side. In the present paper we aim to bridge this gap by pro viding a unified picture of children’s abilities to understand and produce three different types o f pragm atic phenomena - standard com munication acts, deceit, and irony - at three levels - linguistic, extralinguistic, and paralinguistic. W e investigated these abilities in 390 Italian-speaking children aged five to eight years.
W e use Cognitive Pragm atics theory (Bara, 20 10 ) as the theoretical fram ew ork o f the present study. Cognitive Pragm atics theory focuses on the mental processes underlying hum an com munication. It offers a coherent and unified fram ew ork for explaining the com prehension and production o f the different kinds o f pragm atic phenomena investigated, expressed by linguistic, extralinguistic, and paralinguistic means (see also Bara, Bosco & Bucciarelli, 1999a). We used the Cognitive Pragm atics theory, details of which are provided in the next section, as the basis for form ulating our theoretical hypotheses. T h is same theory was also the basis for developing the A ssessm ent Battery o f Com m unication (A B aC o ; Sacco, A ngeleri, Bosco, Colle, M ate & Bara, 2008), parts o f which we used to test the participants in the present study. M ore specifically, we used an experim ental protocol consisting o f three o f the five scales that make up the A B a C o (namely the linguistic, extralinguistic, and paralinguistic scales).
T h e next section provides details o f the experim ental hypotheses concerning the com prehension and production o f different kinds of pragm atic task expressed by linguistic, extralinguistic, and paralinguistic means, that w ill enable us to provide an extensive picture o f the developm ent o f com m unicative abilities in children. Follow in g the tenets o f the Cognitive Pragm atics theory, we argue that the increasing length o f the inferential chain and the increasing com plexity o f the mental representations involved in various types o f pragm atic phenomena explain the different levels o f accuracy with which children com prehend and produce such different phenomena. C hild ren ’ s abilities to deal with inferential chains o f different lengths, as well as with mental representations o f increasing com plexity, develop with age (Bara, B ucciarelli & Johnson- L aird , 1995), and this contributes to explaining the developm ent of their pragm atic competence.
C O G N I T I V E P R A G M A T I C S T H E O R Y
Linguistic and extralinguistic communication
C ognitive Pragm atics (Bara, 20 10 ) is a theory focused on the cognitive processes underlying hum an com munication. Its assum ptions hold for pragm atic phenomena expressed through both linguistic and extralinguistic m eans, i.e., gestures. Cognitive Pragm atics theory assumes that a common com m unicative competence - independent o f the linguistic or extralinguistic gestural means - is instantiated at the level at which a com m unicative intention is form ed or inferred, and com prehended within a specific social context, i.e., at the pragm atic level (Bosco, Bucciarell & Bara, 2004). In an f M R l study, E nrici, Adenzato, Cappa, Bara, and Tettam anti ( 2 0 11 ) showed that a com mon neural network is engaged in com m unicative intention processing independently o f the m odality used. In this section we therefore postulate identical hypotheses for the same pragm atic phenom enon independently o f whether it is produced through language or gesture. In line with such perspective, throughout the paper we have used the term ‘ com m unication acts’ instead o f ‘ speech acts’ , and the term s ‘ actor’ and ‘ p artn er’ instead o f the classical ‘ speaker’ and ‘ listen er’ .
A ccording to Cognitive Pragm atics theory, when two people com m uni cate they are acting on the basis o f a plan, named ‘ behavioral g am e’ , that is at least partially shared by the participants in the dialogue : a behavioral game should be defined as an expected pattern o f interaction shared b y the participants to the dialogue. Consider for exam ple, the following com m unicative exchange :
(1) A N N : “ C ould you take the children this m orning ?”
b e n: “ So rry, I ’m late alre a d y .”
In order to fully understand B en ’ s com m unicative intention, A nn has to recognize the behavioral game she bids through the com munication act, that is in our exam ple :
(2) [B E H A V IO R A L G A M E : F A M I L Y - M A N A G E M E N T ] : • M other or father takes the children to school in the m orning • D epending on the respective engagements, mother and father agree
on who can m ore easily take children to school
A crucial assum ption o f the theory is that the com prehension of the com m unicative m eaning o f any kind o f utterance depends on the recognition o f the behavioral game bid by the actor; participants in a dialogue interpret the com munication acts o f the interlocutor on a basis he/she assumes to be shared between them.
Behavioral games are an indirect m anner o f transm itting the culture in which the infant - and later the child - is im m ersed. F ro m a developm ental perspective, until the knowledge o f a certain behavioral game is acquired
(or sufficiently strengthened), the com m unicative m eaning o f the utterance that makes reference to it cannot be understood (Airenti, 1998).
Length o f the inferential chain: simple and complex communication acts W ithin the theoretical fram ew ork offered b y the Cognitive Pragm atics theory, Bara, Bosco, and Bucciarelli (1999b) proposed abandoning the distinction between direct and indirect speech acts (Searle, 1975) and adopting the distinction between sim ple and com plex com munication acts. D irect and conventional indirect speech acts are called ‘ sim ple com m uni cation a cts’ ; they im m ediately make reference to a behavioral game. Non-conventional indirect speech acts are called ‘ com plex com munication acts’ ; they do not make direct reference to a behavioral game and require a m ore com plex inferential process. T h e authors showed that children from tw o-and-a-half to three years old understand direct speech acts, like ‘ Please close the d o o r’ , and conventional indirect speech acts, such as ‘ W ould you like to sit d o w n ?’ with the same level o f accuracy. B y contrast, they com prehend non-conventional indirect speech acts less accurately, showing difficulty, for exam ple, in understanding that the answer ‘ I t ’ s rain in g’ to the proposal ‘ L e t’ s go out to p la y ’ corresponds to a refusal.
T o sum m arize, the length o f the inferential chain necessary to connect the com munication act to the behavioral game shared between the interlocutors is the factor determ ining the different levels o f difficulty children find in com prehending sim ple and com plex com munication acts. T h is cognitive factor m ay explain w hy, considering the same type o f pragm atic pheno menon, children understand specific utterances, i.e., sim ple com m unicative acts, more accurately than others, i.e., com plex com m unicative acts, even though they are o f com parable sem antic and syntactic com plexity (see also Bosco, Vallana & Bucciarelli, 20 12).
T h e follow ing videotaped interaction is an exam ple o f a sim ple and com plex standard linguistic com munication act investigated in our p ro to col: (3) A le x comes home, S a ra meets him in the doorway.
S a ra says: “ D id you rem em ber to buy m y train tic k e t?” A le x replies...
a. Sim p le: “ Sure, I bought it this m o rn in g .” b. C o m plex: “ Y o u can still count on m y m e m o ry .”
In our exam ple, the standard com munication act (3a) is sim ple because it im m ediately refers to the game [ T O -G O -O N -A N -E R R A N D ] and can therefore im m ediately be interpreted as a positive answer to the actor’s question. B y contrast, in order to com prehend that (3b) is a positive answer it is necessary to infer that people with a good m em ory do not forget to run
their errands. T h u s a longer inferential chain m ust be built in order to com prehend (3a) with respect to (3b).
A s noted by Bosco and B ucciarelli (2008) a further advantage o f the concept o f sim ple and com plex com munication acts is that it also applies to other pragm atic phenomena besides standard com munication. On the basis o f the length of the inferential chain involved, the authors investigated children’ s ability to com prehend sim ple and com plex deceit and irony, showing that, in the linguistic dom ain, children from six-an d -a-half to ten years old com prehend sim ple form s o f deceit and irony more accurately than com plex ones. In the present paper, we extend the difference between sim ple and com plex ironic and deceitful acts to
E X T R A L IN G U IST IC com m unication, which is conceived with respect to the linguistic com munication as an alternative expressive m odality o f the same underlying com m unicative competence (Bara, 2010).
T h e developm ental literature reveals that, as they mature during child hood, children are able to com prehend deceit o f increasing difficulty. C hildren start to use sim ple form s o f deceit, i.e., lies, which are false utterances made with the intention to deceive (Sodian, 19 9 1). L ew is, Stanger, and Sullivan (1989) found that children start to use lies as a means o f escaping a disagreeable consequence from the age o f three years. T h e same com parison can be extended to irony. Children do not find that all form s o f deceit have the same levels o f difficulty (Russell, Jarro ld & Potel, 1995). T h e authors claim that the factor that seems to best explain such diversified perform ance is the increasing cognitive load that the com prehension o f com plex deceit requires.
In line with C ognitive Pragm atics theory, some deceitful com munication acts are sim ple because they consist o f an utterance that denies something that would allow the partner to im m ediately refer to the game the actor wishes to conceal. B y contrast, a com plex deceitful com m unication act consists o f a com m unication act that im plies a belief, which leads the partner to a different game from the game that w ould be reached if the partner had access to the actor’ s private belief. T h e following videotaped interaction is an exam ple o f a sim ple and com plex linguistic deceit we investigated in our p ro to col:
(4) A n drew is eating some biscuits. H e hears K a te arriving, he pushes the empty plate aw ay. K ate sees the empty plate and asks: “ Who has finished m y b iscu its?”
A n drew answers ...
a. S im p le: “ I don’t have the slightest id e a .” b. C om plex: “ I ’m on a d ie t.”
In our exam ple, (4a) is sim ple because it consists o f an utterance which denies the actor’ s private (and true) belief, which im m ediately refers to the
game [ F O O D - S T E A L I N G ] the actor wishes to conceal. Instead, a com plex deceitful speech act, such as (4b), consists o f an utterance which leads to the inference ‘ I f I am on a diet, I cannot eat b iscu its’, which is inconsistent with the game [ F O O D - S T E A L I N G ] the actor wishes to deny. T h u s, the partner needs to process a longer inferential chain to com prehend a com plex deceit.
M oreover, children do not find all form s of irony equally difficult to com prehend. F or exam ple young children are able to com prehend simple form s o f h um or: D unn (19 9 1) analyzed children’s jokes and found that starting from two and three years o f age children are able to understand what their relatives w ill find funny. A s they mature, children learn to p er form more subtle inferences until they reach the levels o f com plex irony. L ucariello and M indolovich (19 9 5), for exam ple, carried out a study on the ability of six- and eight-year-olds to provide ironic endings to unfinished stories. T h e authors suggested that the recognition and construction of ironic events involve the metarepresentational skill o f m anipulating event representations. A ccording to their model, it is possible to make a distinc tion between sim ple and com plex form s o f iron y; their results show that older children construct more com plex ironic endings for the stories from the representational base than younger children do. Fin ally, D ew s et al. (1996) found that young children, contrary to adults, ranked those ironic comments that explicitly state the opposite o f what is meant (direct irony) as funnier than ironic com ments that im ply som ething that is the opposite o f what is said (indirect irony). In our term inology, children com prehend direct irony better than indirect irony because the form er is a simple com munication act and the latter is a com plex one.
A ccording to the speech act approach, in the case o f sim ple irony, the m eaning o f the utterance im m ediately contrasts with a belief shared by the actor and the partner. In particular, in sim ple irony a la G rice (1989), the actor proffers an utterance with the intention o f m eaning exactly the contrary o f what he says. It follows that sim ple irony im m ediately contrasts with the belief given as shared between the interlocutors. On the other hand, com prehension o f com plex irony always involves the detection of its contrast with the shared belief but, in this case, the partner has to go through a com plex inferential chain to detect such a contrast. T h e follow ing videotaped interaction is an exam ple o f the sim ple and com plex extralinguistic irony we investigated in our p ro to col:
(5) H elen is knitting a woolen scarf and realizes she has made a big hole, which she is looking at disconsolately. Sitting in an armchair next to her, is D a v id who ...
a. Sim p le: M akes an O K gesture, as if to say “ W ell d o n e !” b. C o m plex: Looks around him , sees a trophy and gives it to her.
The actor’s communicative intention:
Communication act
is in conflict with his private knowledge?
contrasts with the knowledge given as shared with the partner?
Number o f conflicts
Sincere No No 0
Deceitful Yes No 1
Ironic Yes Yes 2
Fig. I. Conflicts involved com prehending/producing sincere, deceitful, and ironic com munication acts (adapted from Bosco et al., 2009).
O ur exam ple (5a) is a sim ple irony because the ‘ O K ’ gesture im m ediately contrasts with the evidence, shared by the interlocutors, that H elen has not done a good job. On the other hand, (5b) is a com plex irony because it im plies the knowledge that ‘ O nly the best receive tro ph ies’ , contrasting with the evidence shared by actor and partner that H elen has not done a good job.
T o sum m arize, we hypothesized that, on the basis o f the length of inferential chain in vo lved :
Hypothesis i . In both linguistic and extralinguistic communication within each category - standard, ironic, and deceitful acts - children comprehend a simple communication act more accurately than a complex one.
Com plexity o f mental representations : standard, deceitful, and ironic communication acts
W ithin the fram ew ork o f Cognitive Pragm atics theory, Bucciarelli et al. (2003) provided a theoretical explanation based on the increasing com plexity o f the mental representations involved, underlying the com prehension o f standard, deceitful, and ironic com munication acts. T h is cognitive factor, i.e., the com plexity o f mental representation, allows to explain w hy children are more accurate at com prehending com munication acts realizing a specific pragm atic phenomenon with respect to others, having a com parable sem antic and syntactic com plexity.
T h e term ‘ com plexity o f mental representations’ refers to the existence and to the num ber o f conflicts involved in com prehending/producing sincere, deceitful, and ironic com m unication acts (see F igu re i). In standard com m unication, default rules o f inference are used to produce and under stand each other’ s com m unication acts. D efault rules are always valid unless their consequences are explicitly denied (cf. R eiter, 1980). Indeed, in
standard com munication, what the actor says is in Hne with his/her private know ledge: there are no conflicts. Sim ple and com plex com m unication acts (direct and indirect) are all exam ples o f standard com munication. In terms o f mental representations, to generate a standard com munication act the actor has m erely to produce an utterance that is in line with his private knowledge and with the behavioral game s/he shares with the partner. T h u s, in term s o f the com plexity o f the mental representations involved, this is the sim plest case we analyzed. B y contrast, non-standard com m uni cation, such as deceit and irony, involves the violation o f default rules and the occurrence o f conflicting mental representations. It follows that non-standard phenomena are more difficult to produce than standard ones. In addition, among non-standard phenomena, representations involving knowledge expressed by an actor, which is in conflict with knowledge shared with the partner, are m ore difficult to handle than representations that do not involve such a confiict. In particular, in the case o f the production o f a deceitful com m unication act, the actor has to take into consideration the difference between what s/he privately knows and what s/he expresses to the partner. In addition, along with the above-m entioned difference, a statement becom es ironic when the actor also produces a contrast between the expressed mental states and the scenario provided by the knowledge he shares with the partner. T h is makes an ironic utterance the m ost difficult phenomenon to produce, in term s o f the com plexity o f the mental representations and num ber o f confiicts involved (see F igu re i).
T h e existence o f an increasing trend o f difficulty in the c o m p r e h e n s i o n of
standard speech acts, deceit, and irony, has been experim entally validated in children from six-an d -a-half to ten years o f age (Bucciarelli et al., 2003; D e M arco, Colle & Bucciarelli, 2007; Bosco & Bucciarelli, 2008). In the present study, we extend these findings to production, em pirically investigating the existence in children o f an increasing trend o f difficulty in producing standard, deceitful, and ironic com m unication acts in both the
L IN G U IST IC and e x t r a l i n g u i s t i c domains.
A ccording to the literature on the developm ent o f hum an cognition, the ability to detect confiicts between representations is not fully developed in children : it increases with age and correlates w ith the ability to reason (Bara et al., 1995). Based on the assum ptions o f Cognitive Pragm atics theory, along with data from the developm ental literature, we expected to find that the ability to com prehend and produce com munication acts involving con- fiicting mental representations im proves with age. T h u s we expected that:
Hypothesis 2. In both linguistic and in extralinguistic communication there is a trend o f difficulty in the comprehension and production o f different kinds o f pragmatic phenomena, starting from the simplest to the most complex : standard, deceitful, and ironic communication acts.
Hypothesis 3. Children both comprehend and produce standard, deceitful, and ironic communication acts more accurately as their age increases.
Paralinguistic communication
T h e pragm atic m eaning o f com m unication acts also derives from para linguistic elements such as tone, intonation, rhythm , and prosody, which represent sorts o f tributary aspects o f language. F ro m the perspective of C ognitive Pragm atics theory (Bara, 20 10 ), paralinguistic aspects do not have an autonomous status but belong to a connotative dimension. T h e connotative dim ension o f paralinguistic cues integrates com munication acts, highlighting, reinforcing, and eventually m odifying the expressed content, typically in an emotional way.
T h u s, inform ation about the actor’ s feelings can be derived not only from the propositional content, but also from paralinguistic elements such as alternations in his speaking rate, pitch level, pitch contours, and voice quality.
Such emotional features are subsidiary both o f language and of com m unicative gestures. T h e y are parasitical in the sense that they are not autonomous, although by contrast they im prove the effectiveness of com m unication acts, whether expressed using linguistic or extralinguistic m odalities.
Paralinguistic com ponents can be used to express both the actor’ s p ro p ositional attitude and his emotional status. T h e form er refers to the relation between the speaker and the proffered utterance; in this particular context, we have used propositional attitude to indicate the relation between the actor and the com munication act expressed (for exam ple, an assertion im plies a propositional attitude o f belief). T o em pirically investigate the propositional dim ension, we considered four specific types o f standard com munication acts - assertions, questions, requests, and com mands - which correspond to the basic speech acts ( B S A s ; K ash er, 19 9 1). A ccording to K ash er, B S A s are prototypical types o f speech acts and they are generally perform ed b y uttering a specific kind o f sentence, which is linguistically m arked as appropriate for it. In our protocol we investigated B S A s produced using paralinguistic cues and expressing different propositional attitudes.
A s regards emotional status, we considered items in which paralinguistic elements transm it emotional conditions. W e focused only on basic emotions such as anger, fear, happiness, and sadness. T h e developm ental literature has shown that even prelinguistic listeners are sensitive to the emotional tinge o f com m unicative interaction (Ekm an & Oster, 1979). Infants can recognize emotional facial expressions early on in life and regulate their behavior accordingly (Striano & Vaish, 2006), and by three to four years o f age are able to recognize and name emotions on the basis o f various
expressive cues (Pons, H arris & de R onsey, 2004). Furtherm ore, studies have shown that children’ s ability to interpret speaker’ s affect from para- linguistic cues (Friend & Bryant 2000), and to recognize facial emotional expressions (H erba & Phillips, 2004), im proves with age.
T o the best o f our knowledge, nobody has investigated emotional aspects from the perspectives o f both com prehension and production, and nobody has com pared the ability to recognize emotion with the ability to recognize paralinguistic aspects, which also transm it propositional attitudes. In the present study we aim to assess such abilities on a global basis, in order to obtain a com prehensive picture o f the pathw ay o f developm ent o f para linguistic com ponents after the preschool period. We also wish to explore the developm ent o f paralinguistic aspects linked to propositional attitudes vs. emotional content, in order to further understand how these phenomena interact (for a detailed description o f the tasks see the ‘ M aterial : Paralinguistic scale’ section). In particular, we expect the results to reflect the fact that children master the ability to recognize and generate emotions earlier than the capacity to understand and follow social rules.
Hypothesis 4. Children recognize and produce paralinguistic aspects referring to emotional states more accurately than paralinguistic aspects referring to propositional attitudes.
Furtherm ore, paralinguistic elements are usually in line with the expressed content proffered, reinforcing the same meaning. H ow ever, in everyday conversation there is sometime a m ism atch between paralinguistic indicators and expressed content: for exam ple, a wom an could say ‘ So nice to see you here a gain ’ with a cold tone and attitude to the ex-husband she hates and has met by chance. W e have labeled this phenomenon ‘ paralinguistic contradiction’ , in order to indicate those situations in which paralinguistic cues reveal mental states different from the expressed sem antic content.^
W e introduced this com m unicative case in our experim ental protocol in order to verify whether children pay more attention to verbal content or to paralinguistic cues when these are conflicting. In children, once language acquisition is under w ay, the principal source o f inform ation about other people’ s feelings becomes the linguistic input, m ostly because they learn the speciflc m eaning o f w ords, in particular emotional words referring to mental states (e.g., Bretherton, Fritz, Zahn-W axler & R id gew ay, 1986). F rom this
[i] T h e main difference between paralinguistic contradiction and irony or sarcasm is the actor’s com m unicative intention. In proffering an ironic or sarcastic utterance the actor wishes the partner to recognize the contrast between the mental states he/she expresses and his/her private belief. In the paralinguistic contradiction the actor has no such com m unicative intention.
stage forward, children seem more sensitive to the expressed content rather than to paralinguistic cues (Ross, 2000). In a study investigating children’ s understanding of emotion in speech, M orton and T reh u b (2001) used a protocol where cues conflicted (i.e., a happy situation was described with sad paralinguistic elements), and they found that children relied prim arily on the expressed content: the children could interpret the paralinguistic cues but they sim ply accorded greater weight to the propositional elements until eight years o f age, whereas older children and adults considered all available cues and relied prim arily on paralinguistic aspects (M ehrabian & W iener, 1967). In particular. F riend and Bryant (2000) showed that, when paralinguistic cues and sem antic content are discrepant, children of four and seven years o f age favor the interpretation based on semantic content, whereas children o f ten years o f age favor the interpretation based on paralinguistic cues (for a detailed description o f the task see the ‘ M aterial - Paralinguistic scale’ section).
T h u s, we hypothesize that:
Hypothesis 5. In the case o f paralinguistic contradiction, younger children focus their attention more on linguistic content than on paralinguistic cues whereas older children focus their attention more on paralinguistic cues than on linguistic content.
Fin ally, in line with the relevant literature, we expected that:
Hypothesis 6. C hildren’s ability to master paralinguistic aspects referring to an emotional state, propositional attitude, and paralinguistic contradiction increases with their age.
S O C IO E C O N O M IC S T A T U S A N D P R A G M A T I C D E V E L O P M E N T
F am ily socioeconomic status (S E S ) is a predictor o f m any aspects o f child developm ent, particularly language developm ent: the children o f more educated and advantaged parents have greater vocabulary skills and faster vocabulary grow th than those o f less educated and advantaged parents (e.g.. R ow , 2008). F ew studies have investigated whether socioeconomic aspects can also affect the developm ent o f social-pragm atic skills. H ow ever, some differences have been reported, for exam ple h ig h -S E S parents seem to m ore often verbally encourage and provide feedback to their children than lo w -S E S parents (H art & R isley, 1995).
In the present study, for explorative purposes, we examine the relation between S E S and children’s pragm atic developm ent after the preschool period, in order to identify some possible influences o f S E S on pragm atic perform ance. T o the best o f our knowledge, no study has investigated the
relation between S E S and pragm atic ability during m iddle ch ildh ood: S E S im pact on language skills during early childhood seems robust, and the same pattern has em erged in studies focused on academic achievement (for a critical review see W hite, 1982). Pragm atic perform ance represents a com plex skill, which requires a wide range o f cognitive abilities - i.e., attention, m em ory, planning, and T h eo ry o f M in d (T irassa, Bosco & Colle, 2006a, 2006b); since several studies have shown that fam ily S E S is associ ated with a wide array o f cognitive outcomes in children (Bradley & Corw yn, 2002), the role o f S E S should also be present in pragm atic development. H ow ever, S E S effects play a role at m ultiple levels and are mediated by the children’s personal disposition, fam ily characteristics, and external resources, such as the m ultiple contexts that children experience during the schooling period.
Hypothesis 7. We explore the influence o f S E S in all the pragmatic phenomena investigated, expecting a moderate overall effect on children’s performance.
T H E P R E S E N T S T U D Y
T o sum m arize, the goals o f the present are: (i) to provide a com prehensive picture o f children’ s pragm atic abilities; and (ii) to outline a theoretical explanation for the developm ental pathw ays observed. T o achieve these goals we form ulate a series o f experim ental hypotheses :
1 . In both linguistic and extralinguistic com m unication within each category - standard, ironic, and deceitful acts - children com prehend a sim ple com m unication act more accurately than a com plex one.
2. In both linguistic and in extralinguistic com m unication there is a trend of difficulty in the com prehension and production o f different kinds of pragm atic phenomena, starting from the sim plest to the m ost com plex : standard, deceitful, and ironic com munication acts.
3. Children both com prehend and produce standard, deceitful, and ironic com m unication acts more accurately as their age increases.
4. Children recognize and produce paralinguistic aspects referring to emotional states m ore accurately than paralinguistic aspects referring to propositional attitudes.
5. In the case o f paralinguistic contradiction, younger children focus their attention more on linguistic content than on paralinguistic cues whereas older children focus their attention m ore on paralinguistic cues than on linguistic content.
6. C hild ren ’s ability to m aster paralinguistic aspects referring to an emotional state, propositional attitude, and paralinguistic contradiction increases with their age.
7- W e explore the influence o f S E S in all the pragm atic phenomena investigated, expecting a moderate overall effect on children’ s perform ance.
W e explore these hypotheses by adm inistering the Lin gu istic, Extralinguistic, and Paralinguistic scales, which are part o f the Assessm ent Battery for Com m unication (A B aC o ; Sacco et al., 2008; Bosco, A ngeleri, Zuffranieri, Bara & Sacco, 2 0 12 ), to 390 Italian-speaking children divided into three age groups from flve through eight years o f age. Cognitive Pragm atics theory (Bara, 20 10 ) has been the basis for developing A B a C o , and we also use this theory to develop our hypotheses.
M E T H O D
Participants
A total o f 390 children aged flve to eight years participated in the study. We investigated these abilities in children aged flve to eight because previous research showed that this is the age range in which children are developing the relevant skills (Bucciarelli et al., 20 03; Bosco & Bucciarelli, 2008). T h ey were divided into three age groups: 13 0 children ranging from 5 ;o to 5 ;6 { M = 5 ;3 , S D = 2 months), 13 0 children ranging from 6 ;6 to 7 ;o ( M = 6 ;8 , S D = 3 months), and 13 0 children ranging from 8 ; o to 8 ; 6 { M = 8 ; 2, S D = 2 months). W ithin each age group, there were an equal num ber o f males and females. Children were recruited through public and private nursery and elem entary schools in the T u rin area o f northern Italy. D uring an initial visit to the schools, a research assistant described the study in detail to the teachers, who then selected the children according to their age and sex. A letter was sent to the parents o f the children deemed suitable to take part in the study, together with the inform ed consent form , which they were asked to flll in. T h e children whose parents gave their consent were in cluded in the sample. A ll were Italian native speakers. Socioeconom ic status was m easured by fam ily com position, parental educational level, and occu pation, and was obtained using a questionnaire fllled out by the participants’ parents. T h e S E S index was derived from the T w o -F a cto r Index of Social Status (H ollingshead, 1975);^ we updated the em ploym ent categories
[2] T o calculate the S E S score o f the children’ s fam ilies, scale values for occupation (ranging from I to 7) and for education (ranging from i to 7) o f both parents were m ultiplied by factor weights o f 7 and 4, respectively. T h ese two products were then summed, providing the Index o f Social Position (IS P ). I f both parents were em ployed, the mean o f their indexes was calculated; if ju st one o f them was em ployed, her/his index was used. Possible scores on I S P ranged from 77 (lowest social class) to 1 1 (highest social class). W e used the continuous I S P values to evaluate the S E S contribution to children’ s pragm atic perform ance, while to better describe the experim ental sample we classified the values into five groups (i.e., lower, lower-m iddle, m iddle, upper-m iddle, and upper) follow ing H ollingshead’ s suggestions.
included in the H ollingshead procedure with reference to the current Italian social context. T h e highest percentage o f children were from the m iddle social class (39-5% ), but they were also distributed among the other classes (low er: 8-7% ; low er-m iddle: 2 1 % ; and upper-m iddle: 23-6% ). M ore detailed inform ation about fam ily com position and socioeconomic factors integrating the inform ation on S E S with age are shown in T ab le i .
M ateria l
T h e A ssessm ent Battery for Com m unication (A B aC o) is a clinical instrum ent for the evaluation o f pragm atic abilities in patients affected by neurological or psychiatric disorders involving com munication im pairm ents. It has been developed for both adults and children: some items were thus created in two different form s. E xp erts’ judgm ent provided an independent validation o f the content o f the battery and confirmed its suitability for both children and adults. A B a C o showed the validity of the underlying theoretical construct, high inter-rater agreement, and good internal consistency (Sacco et al., 2008; A ngeleri, Bosco, G abbatore, Bara & Sacco, 2 0 12 ; Bosco, A ngeleri, Zuffranieri, Bara & Sacco, 20 12).
T h e experim ental protocol consists o f three o f the five scales com posing the A B a C o : (i) Lin gu istic, (2) Extralinguistic, and (3) Paralinguistic.
T h e whole protocol included five evaluation scales for a total o f 88 items : 16 in vivo items based on the exam iner’ s prom pts and 72 items based on videotaped scenes. Each evaluation scale was subdivided into com prehen sion and production sections (see ‘ A ppendix A ’ for a description o f the structure o f the experim ental protocol). T h e videotaped scenes observed by participants were perform ed and recorded for the purpose o f the present study. Each lasted 2 0 -2 5 seconds, and com prised a controlled num ber of words (range: 7 ± 2 ) . T h e scenes were designed to represent com m unicative interactions between two people. In com prehension scenes, the participant saw a complete interaction (i.e., actor A produces a com m unicative act and partner B replies). In production scenes, the participants saw only the initial part o f the interaction (i.e., actor A produces a com m unicative act) and were asked to respond appropriately. T o better illustrate, we provide an exam ple o f a ‘ sim ple standard - extralinguistic com prehension’ scene :
(6) Francesco is talking on the phone. Luisa comes in and makes a gesture as i f to say : ‘ Shall we go ? ’ Question to the participant : ‘ In your opinion, what could the boy answer to the girl ? ’
Other exam ples o f items, and participants’ responses, are reported in ‘ A ppendix B ’ .
T A B L E I . F am ily composition and socioeconomic details o f children’s sample
A ge group 5 ; 0 -5 ; 6 A ge group 6 ; 6 -7 ; 0 A ge group 8 ; 0-8 ; 6 G lobal sample M ean (SD ) N % M ean (SD ) N % M ean (SD ) N % M ean (SD ) N %
Fam ily composition
Both parents - 1 1 7 90 - I I I 85-4 - 1 1 2 86-2 - 340 87-2 Single parent - 6 4-6 - 8 6-2 - 8 6-2 - 22 5-6 O nly child - 3 1 23-8 - 36 27-7 - 23 17-7 - 90 23-1 N ot only child - 92 70-8 - 83 63-8 - 97 74-6 - 272 69-7 N ° o f sisters/brothers •93 (-6) - - •97 ( i- i) - - I-O (-7) - - •96 (-8) - -S E -S '
M other’ s education 1 1 ; 1 1 yrs (3 ; 6) - - 1 2 ; 2 yrs (3; 6) - - I 2 ; 4 yrs (3 ;s) - - 1 2 ; 2 yrs (3; 6) -
-Father’ s education i i ; i i yrs ( 3 ;s ) - - I 2 ; 4 yrs (3 ;s) - - 1 2 ; I yrs (3 ; 8) - - 1 2 ; I yrs (3; 6) -
-Fam ily ISP^ 43'5 (i2 '8 ) - - 40-5 (13-2) - - 40-9 (12-5) - - 41-6 (i2-8) -
-Category
L ow er - 14 10-8 - II 8-5 - 9 6-9 - 34 8-7
Low er-m iddle - 35 27 - 24 18-5 - 23 17-7 - 82 2 1
M iddle - 50 38-5 - 49 37-7 - 55 42-3 - 154 39'5
U pper-m iddle - 24 i 8'5 - 35 26-9 - 33 25-4 - 92 23-6
N o information - 7 5-4 - II 8-5 - 10 7-7 - 28 7-2
N O TES :
' S E S = Socioeconom ic status. ^ I S P = Index o f Social Position.
In the follow ing sections we describe the three evaluation scales and the specific pragm atic phenomena included in each one.
Linguistic and extralinguistic scales
T h e linguistic scale included tasks aimed at investigating the com prehen sion and production o f com munication acts expressed prim arily through linguistic means. T h e extralinguistic scale also assessed the com prehension and production o f com munication acts, but expressed only through extralinguistic means. Since the two scales include the same com munication acts, they are described together.
T h e pragm atic phenomena investigated here are standard com munication acts, deceit, and irony (both sim ple and com plex). T h e subject watched short videos where two characters were engaged in a com m unicative interaction.
Comprehension. T h e actor asked his/her partner a question and the partner replied. T h e subject had to understand the com munication act produced by the partner in reply. In the linguistic scale, the characters com m unicated verbally, whereas in the extralinguistic scale, they com m unicated through gestures alone.
Production. In this case the actor said something to the partner and the video stopped. T h e subject was requested to answer the actor, assum ing the partner’ s perspective. In the linguistic scale, the com m unicative interaction occurred in the linguistic m odality and the subject had to reply verbally. In the extralinguistic scale, the actor perform ed com m unicative gestures without any language support and the subject had to reply using gestures alone.
Paralinguistic scale
T h e paralinguistic scale included the com prehension and production of those aspects that generally accom pany a com munication act, such as proxem ics and prosody.
Comprehension. T h e pragm atic phenomena investigated here are basic speech acts ( B S A s ; K ash er, 19 9 1 : assertions, questions, requests, and commands), basic emotions (anger, fear, happiness, and sadness), and paralinguistic contradiction.
Basic speech a cts: the exam iner showed the subject short videos where an actor, speaking an invented language, makes an assertion, asks a question, makes a request, or gives a command. T h e subject had to understand the type o f com munication act proffered by the actor, focusing only on paralinguistic indicators, since the language was purposely sem antically meaningless. T h e exam iner provided four possible answers, only one of
which was correct. F o r exam ple, the actor in the scene gave a com mand and the subject was requested to choose from among the follow ing altern atives: ‘ T h e actor w ants: (i) to make a request (2) to say som ething that he believes true (3) to give a com mand (4) to jo k e .’®
Basic em otion s: the exam iner showed the subject short videos where an actor, once again speaking in an invented language, expressed a basic emotion. T h e subject had to recognize the emotion, focusing on the paralinguistic indicators. T h e exam iner provided four possible answers, only one o f which was correct. F or exam ple, the actor expressed anger using paralinguistic elements and the subject was requested to choose the correct answer from among the fo llo w in g : ‘ In your opinion, how does the actor feel? (i) sad, (2) angry, (3) happy, (4) sca red .’
Paralinguistic contradiction: the exam iner showed the subject short videos in which two characters were engaged in a com m unicative inter action : one o f the actors verbally expressed something that is in contrast with the paralinguistic indicators (i.e., the actor says ‘ I like this present very m u c h !’ while his voice and attitude reveal that he does not like it at all). T h e subject had to grasp the inconsistency between expressed content and paralinguistic indicators, saying, for instance, in the exam ple that the actor really does not like that present.
Production. T h e pragm atic phenomena investigated here are basic com m unication acts and basic emotions.
Basic speech a c ts: the exam iner asked the subject to produce assertions, questions, requests, and com mands, paying special attention to the paralinguistic indicators. F or exam ple, the exam iner told the subject to ‘ Ask me whether it is sunny to d a y ’ or ‘ T e ll me that it is sunny to d a y ’ .
Basic em otions: the exam iner asked the subject to produce com m uni cation acts colored b y a specific emotion or m ood; the exam iner provided the sem antic content o f the requested act and the emotion with which it has to be expressed. F o r exam ple, the exam iner asked the subject to ‘ T e ll me that you have received a letter. T e ll me that in a happy w a y .’
Content validity
In order to assess whether the items o f our protocol were appropriate for children ranging in age from five to eight-and-a-half years old, four developm ental psychologists (authors o f at least one international publication focused on child development) were recruited. T h e experts
[3] T h e children were very fam iliar with all the terms used in the tasks, and did not appear to have any difficulty in understanding these alternative options. T h e translation from Italian to English m ay contain terms with which English-speaking children m ight not be fam iliar. T h is could give the impression that the children taking part in the study had some difficulties, whereas in actual fact they did not.
were given an electronic form to fill in individually, in which they were required to evaluate each item included in the experim ental protocol. F or each item, they had to rate a statement o f appropriateness, focusing on item suitability for children aged 5 ;o to 8 ;6 , on a five-point L ik ert scale. T h e mean value was 4-9 {S D = -i), showing that the experts ju dged the items to be adequate.
Procedure
T h e children com pleted the protocol tasks individually with one o f the authors or a research assistant in a quiet room at their school. T h e y were video-recorded during the experim ental sessions, in order to enable post test scoring. T h e average time required to adm inister the entire protocol ranged from 45 to 60 minutes. T w o independent ju dges rated the children’ s responses; the raters were blind to the aims o f the present research.
W e divided the experim ental protocol into two versions (A and B), each containing the same num ber o f items. T h e items were randomized differently in the two versions and sim ple and com plex com prehension items were allocated equally. Each standard and non-standard scene was recorded in two versions : given the same context and the same characters, one version showed a com m unicative interaction in which the partner answered the actor with a sim ple com munication act, whereas the second one showed a com plex com m unication act (see exam ples in ‘ A ppendix B ’ ). T h u s each version o f the protocol (A or B) contained only one version (simple or complex) of the same scene; the total num ber o f sim ple and com plex scenes was balanced in protocols A and B. H a lf o f the participants dealt with protocol A , while the other h alf dealt with protocol B. Each child was random ly assigned to protocol A or B.
Coding procedure
Scoring was kept on specific score sheets by two independent judges, while watching the children’ s video-recorded experim ental sessions. F o r each task it was possible to obtain a score o f o or i , depending on whether the answer was correct (i point) or incorrect (o points) (see exam ples in ‘ A ppendix B ’ ). T h e level o f agreement for the ratings assigned by the two independent judges was calculated using C ohen’ s kappa coefficient. T h e kappa values could range between o and i . K ap p a values o f -zo or less indicate a slight com pliance, -2i to -40 a fair agreement, -41 to -6o a moderate agreement, -6i to -8o a substantial agreement, and ^ -8o an almost perfect agreement. C ohen’ s kappa was calculated for 60 participants, representing m ore than 15 % o f the total sam ple. T h e kappa coefficient was -95, indicating an almost perfect agreement (Lan dis & K o ch , 1977).
A ll discrepancies were resolved by a discussion in order to reach a final resolution accepted by both coders.
R E S U L T S
Linguistic and extralinguistic scale W e hypothesized that:
• In both linguistic and extralinguistic com munication within each category - standard, ironic, and deceitful acts - children would com prehend a sim ple com munication act more accurately than a com plex one (H ypothesis i).
• In both linguistic and in extralinguistic com munication there would be a trend o f difficulty in the com prehension and production of different kinds o f pragm atic phenomena, starting from the sim plest to the most co m p lex: standard, deceitful, and ironic com m unication acts (H ypothesis 2).
• Children would both com prehend and produce standard, deceitful, and ironic com m unication acts more accurately as their age increases (H ypothesis 3).
W e tested all three hypotheses together using one A N O V A to analyze the com prehension results on the linguistic and extralinguistic scales, and a second A N O V A to analyze the production results. In reporting our results we first o f all consider those concerning the com prehension o f the linguistic and extralinguistic scales and then those concerning the production o f the linguistic and the extralinguistic scales.
Comprehension o f simple and complex communication acts: standard, deceitful, and ironic
T o investigate children’s perform ance in the com prehension o f sim ple and com plex standard, deceitful, and ironic com munication acts, data were entered into the A N O V A with one betw een-participants factor (a g e g r o u p,
with three levels, corresponding to the three different age groups) and two w ithin-participants factors (t y p e o f p h e n o m e n o n, with three le v e ls: standard, deceitful, and ironic; and t y p e o f i n f e r e n c e, with two levels: sim ple and com plex), for both the linguistic and extralinguistic scales. A Bonferroni correction o f a/3 was applied for the post-hoc pairwise com parisons between age groups. D ata are shown in F igu res 2 and 3.
In the com prehension o f the linguistic scale, the A N O V A revealed a main effect o f the type o f inference ( i ’(i_364)= i4 '9 ; P < -o o o i; 7}^= -04): children in all three age groups understood sim ple com munication acts more accurately than com plex ones. A i-test analysis revealed that the difference between
I Simple ' Complex
Standard Deceit Irony Standard Deceit Irony
L i n g u is t i c E x t r a l i n g u is t ic
Fig. 2. C hild ren’ s com prehension o f linguistic and extralinguistic com munication acts (global sample). T h e spotted line indicates the general trend o f difficulty o f the different pragm atic phenomena.
15:0-5 ;S
■e;0-S;S
Standard Deceit Ironv Standard Deceit Ironv
Linguistic Extralinguistic
Fig. 3. Percentage o f correct answers obtained b y the different age groups in the com prehension o f standard, deceitful, and ironic com munication acts.
sim ple and com plex com munication acts was significant only in the case o f standard acts and deceit (i-test: 2 - 5 9 < i < 4 '8 ; -oooi </) < -oi), whereas the diff'erence between sim ple and com plex irony was not significant (i-test : i = -2; p = -84, see F igu re 2).
In all three age groups there was also a main efi'ect o f the type o f p rag m atic phenomenon (-F’(2_72g) = 267-57; p < - o o o i', r f = -/\.z). W e introduced a linear contrast to test whether there is a linear increase (or decrease) in the scores; the analysis revealed a linear decrease depending on the type of pragm atic phenomenon ( i ’(i,364) = 11 7 ; /><'OOOi; r f = -z/^): contrary to our expectation children overall and within each age group understood deceit m ost accurately, followed by standard com m unication acts and lastly irony, which they understood least accurately (see F igu re 3).
T h e A N O V A also revealed a m ain effect o f the age group (-F’(2_364) = 80-87 ; p < - o o o i\ indicating that children’ s perform ance im proved with age in all tasks. Post-hoc analysis using the Bonferroni m ultiple com parison correction revealed that the difference in perform ance was significant in each pair o f age groups (p < -o o o i).
D ata concerning the com prehension o f the extralinguistic scale were also entered into the A N O V A , with a g e g r o u p as the betw een-participants
factor. T h e A N O V A revealed a main effect o f type o f inference (-F’(i 3oe) = i0 2 - i; /> < -o o o i; = in extralinguistic comprehension, children under stood sim ple com m unication acts m ore accurately than com plex ones in all pragm atic phenomena (i-test: 3 - 8 7 < i < 1 1 -46; p < - o o o i, see F igu re 2).
T h ere was also a main effect o f the type o f pragm atic phenomenon (^(2,612) = 15 6 -3 7 ; /> < -o o o i; )7^ = -34). T h e linear contrast revealed a linear
decrease in scores depending on the type o f pragm atic phenomenon (^(1,306) = 3 3 3 '5 2 ; p < - o o o i ; rj^ = 'S)'- children overall and within each age group understood standard com munication acts most accurately, followed by deceit and irony, which they understood least accurately.
W e also found a m ain effect o f the age group ( i ’(2,3oe) = 6 3 'i ; p < - o o o i\ = perform ance im proved with the children’ s age. A post-hoc com parison revealed that the difference in perform ance was significant in each pair o f age groups (-oooi < p < - o o i , see F igu re 3).
Production o f standard, deceitful, and ironic communication acts
T o investigate children’s perform ance in the production o f standard, deceitful, and ironic com m unication acts, data were entered into the A N O V A with one betw een-participants factor (a g e g r o u p, with three levels, corresponding to the three different age groups) and one within- participants factor (t y p e o f p h e n o m e n o n, with three levels : standard, deceitful, and ironic), for both the linguistic and extralinguistic scales.
T h e A N O V A revealed a pattern o f differences among types o f pragm atic phenomena, both in the linguistic scale (^’(2,742) = 8 3 3 '4 7 ; p < - o o o i ; yf‘ ='~l) and extralinguistic scale (^’(2,702) = 46 5'55 ; p < -oooi ; rj^=-6). In more detail, we found the follow ing trend o f difficulty in linguistic production : standard com munication acts were produced m ost accurately, followed by deceit and finally by ironic com munication acts, which were produced least accurately (Lin ear C ontrast: i ’(i_37i) = 1 367-01 ; p < - o o o i ; r f =■?>). F o r the extralinguistic scale, the linear contrast revealed the same trend o f difficulty (■P’(i,36i) = 6i3'7 5; / ’ < -o o o i ; »7^ = -6, see F igu re 4).
F o r the linguistic scale, the A N O V A revealed a pattern o f differences among age groups (^’(2,371)= i7 ' 3 ; /> < -o o o i; r f = -i). Post-hoc com parison showed significant differences in perform ance between the following age grou ps: 5 ;0 - 5 ;6 vs. 6 ;6 - 7 ;o (/><-oooi) and 5 ;0 - 5 ;6 vs. 8 ;0 - 8 ;6 (/)< -o oo i), whereas the difference between the 6 ;6 - 7 ;o and 8 ;0 - 8 ; 6 age groups was not significant (p =
-(>'i)-In the extralinguistic scale, we found the same trend in differences among age groups ( i ’(2,36i) = 7 9 ' 4 7; /> < -o o o i; A lso, in this case, post-hoc com parisons revealed significant differences in perform ance between the 5; 0 - 5 ; 6 vs. 6 ;6 - 7 ;o and 5; 0 - 5 ; 6 vs. 8; 0 -8 ; 6 age groups (/><-oooi), whereas the difference between the 6 ; 6 -7 ; o and 8 ; 0 -8 ; 6 age groups was not significant (p = - ii) . T h ese data are shown in F igu re 5.
Paralinguistic scale
W e hypothesized that children would recognize and produce paralinguistic aspects referring to emotional states m ore accurately than paralinguistic aspects referring to propositional attitudes (H ypothesis 4).
100
80 1 60 40 ■20
-Standard Deceit LinguisticIrony Standard Deceit Extralinguistic
Irony
Fig. 4- Percentage o f correct answers obtained b y children in the production o f linguistic and extralinguistic com munication acts (global sample). T h e spotted line indicates the general trend o f difficulty o f the different pragm atic phenomena.
I 5 :0 -5;6
■ 6;6-7;0
■ 8;0-9;6
Standard Deceit Ironv
L in g u istic
Standard Deceit Ironv
E x tra lin g u is tic
Fig. 5- Percentage o f correct answers obtained b y the different age groups in the production o f standard, deceitful, and ironic com munication acts.
W e also expected that in the case o f paralinguistic contradiction, younger children w ould focus their attention more on linguistic content than on paralinguistic cues whereas older children would focus their attention more
T A B L E 2 . M ean {SD ) o f children’s performance on paralinguistic scale {range o - i)
Basic
A ge group com municative acts Basic emotions Contradiction Paralinguistic
Com prehension 5 ; o - 5 ; 6 -27 (-i8) -85 (-19) -38 (-29) 6 ; 6 - 7 ;o -46 (-24) -92 (- 13 ) -69 (-29) 8 ; o - 8 ; 6 -55 (-24) -93 (-12) -87 (-22) O verall -43 (-25) -go {-15) -65 {-33) Paralinguistic Production 5 ;o ~ 5 ;6 "47 ('29) '4 1 ('32) — — 6 ; 6 - 7 ;o -70 (-28) -76 (-26) 8 ; o - 8 ; 6 -83 (-2) -84 (-19) O verall -67 (■3) -67 (■3)
-nil paralinguistic cues than on linguistic content (H ypothesis 5). F in ally we expected that children’s ability to master paralinguistic aspects referring to an emotional state, propositional attitude, and paralinguistic contradiction would increase with their age (H ypothesis 6). C hild ren ’ s perform ance on the paralinguistic scale is displayed in T ab le 2.
T o investigate children’ s perform ance in the com prehension of paralinguistic aspects, data were entered into the A N O V A with one betw een-participants factor ( a g e g r o u p , with three levels, corresponding to the three different age groups) and one w ithin-participants factor ( t y p e OF PHENOMENON, with two levels: basic emotions and basic com munication acts). T h e A N O V A analysis revealed a significant difference among age groups (-F’(2_367) = 56-67; p < - o o o i\ r f = '24) in the com prehension o f basic emotions, and basic com m unication acts; the post-hoc com parison revealed significant differences in perform ance between each pair o f age groups {p < -o o o i). Children in all three age groups understood basic emotions more accurately than basic com munication acts (-F’(2_367)= 126 0-5; p < - o o o i ;
t]^ = -77 ).
T o investigate perform ance in the p r o d u c t i o n o f paralinguistic elements,
we conducted an A N O V A with one betw een-participants factor (a g eg r o u p)
and one w ithin-participants factor (t y p e o f e l e m e n t, with two levels : basic com munication acts and basic emotion). T h e analysis revealed a sig nificant difference among the age groups (-F’(2_367) = 30-6; p < - o o o i ; rj^ = -4i). T h e post-hoc com parison revealed significant differences in perform ance between the 5; 0 - 5 ; 6 vs. 6 ;6 ~ 7 ;o age groups and the 5; 0 - 5 ; 6 vs. 8; 0 -8 ; 6 age groups {p < -o o o i ), whereas the difference between the 6 ;6 ~ 7 ;o and 8 ;0 - 8 ;6 age groups was not significant {p = -i6). T h e analysis revealed no main effect o f type o f element (_F(j306) = 2-46; p = - iz ; rj^ = -02).
