Emotional representation in facial expression and script A comparison between normal and autistic children

Transcription

1 Research in Developmental Disabilities 28 (2007) Emotional representation in facial expression and script A comparison between normal and autistic children Michela Balconi *, Alba Carrera Laboratory of Cognitive Psychology, Department of Psychology, Catholic University of Milan, Largo Gemelli 1, Milan, Italy Received 9 February 2006; received in revised form 6 April 2006; accepted 22 May 2006 Abstract The paper explored conceptual and lexical skills with regard to emotional correlates of facial stimuli and scripts. In two different experimental phases normal and autistic children observed six facial expressions of emotions (happiness, anger, fear, sadness, surprise, and disgust) and six emotional scripts (contextualized facial expressions). In the second place, the effect of emotional domain (different emotions) in decoding was explored. A semantic grid was applied to conversational line, including two levels of data: the lexical adequacy index (correct decoding of emotion) and the emotional vocabulary (such as the causal representation and the hedonic valence of the stimulus). Log linear analysis showed different representations across the subjects, as a function of emotion, task and pathology. Specifically, childrens lexical competence was well developed for some emotions (such as happiness, anger, and fear), and as a function of type of task, that is script was better represented than face. Between the main linguistic indexes, causal relation was a prototypical index for emotional conceptualization. Finally, pathology affected children s performance, with an increased facilitation effect for autistic children in the script condition. # 2006 Elsevier Ltd. All rights reserved. Keywords: Facial expressions; Autism; Emotion; Script 1. Introduction In the last two decades, developmental psychology has seen an increasing interest in the emotion comprehension. Emotional face recognition and understanding represents a primary social competence, because it contributes to social interactions and social management (Balconi * Corresponding author. Tel.: ; fax: address: michela.balconi@unicatt.it (M. Balconi) /$ see front matter # 2006 Elsevier Ltd. All rights reserved. doi: /j.ridd

2 410 M. Balconi, A. Carrera / Research in Developmental Disabilities 28 (2007) & Lucchiari, 2006). Several researchers suggested that autistic children present a dysfunction in this emotional domain (Kanner, 1943). They are characterized by an impairment of the ability to express their emotional internal states and to decode and understand others emotion. Moreover, they show an inattentiveness and an apparent indifference towards the face of other people. Indeed, several studies (Celani, Battacchi, & Arcidiacono, 1999) have found that, for autistic children, facial emotions have a weak salience, if compared with other non-emotional cues. They tend to ignore emotional expressions, unless they are explicitly required to do it. Nonetheless, the extent of the deficit in facial expression understanding varies (Gepner, de Gelder, & de Schonen, 1996). In an experimental review about face recognition in subjects with mental retardation, Rojahn, Lederer, and Tassé (1995) observed that, when the cognitive functioning level diminishes, the ability to decode emotions from facial expression diminishes. Even in the more specific case of autism, it is attested that the developmental level plays a central role in determining the ability to decode and understand facial expression of emotions. Low-functioning autistic individuals present an underdeveloped decoding ability, while some high-functioning subjects show a level of performance similar to normals (Rojahn et al., 1995). An overview of the available literature allows us to conclude that, autistic people perform worse than normals, but how great is this impairment? This difficulty suggests the presence of a circumscribed and specific deficit, weakening the hypothesis of a damage of the whole ability to decode the emotional face (Bormann-Kischkel, Vilsmeier, & Baude, 1995). Another main factor related to decoding competences is the type of emotions they have to recognize (Bormann-Kischkel et al., 1995). Autistic subjects shown to be competent in the decoding of the primary or simple emotions (e.g. happiness and sadness), show more difficulties in processing secondary or complex emotions, such as pride and embarrassment (Balconi & Lucchiari, 2005; Balconi & Pozzoli, 2003a; Capps, Yirmiya, & Sigman, 1992). To identify these emotions more time and more informational cues must be analyzed. Moreover, as regard to the secondary emotion, a more accentuated difficulty in understanding causal antecedents (the events that caused the emotion expressed by face) and contextual relations (the social situations in which the emotion is produced) emerges (Hillier & Allinson, 2002). Bormann-Kischkel et al. (1995) observed a specific difficulty in understanding the emotions that present a lack of correspondence between people expectations and environment events. These emotions have an external and social origin, such as surprise, dismay, and astonishment. In parallel, Capps et al. (1992) observed a greater impairment in recognizing the expression of those emotions that have an external locus of control and, simultaneously, that require a wide knowledge of the social scripts and of their social consequences. Bormann-Kischkel, Amorosa, and von Benda (1993) suggested that the comprehension of emotional expressions that are explained by external events constitutes a precursor of the theory of mind. In line with this hypothesis, Baron-Cohen, Spitz, and Cross (1993) suggested that the comprehension is more difficult for emotions that imply the activation of some cognitive functions, such as mentalization and metarepresentation. Another main concern is represented by contextual and situational elements that cause emotion (Fein, Lucci, Braverman, & Waterhouse, 1992). Therefore, it is necessary to take into account the role of a wider socializing context. Emotion recognition is allowed by the development and the generalization of an emotional script, that is, a child can recognize a specific emotion by verifying the presence of several prototypical elements that are arranged in precise causal and temporal sequences. These scripts include not only facial expressions, but also the representation of causal factors, physical and social context, several actions and their consequences, as well as the cognitive appraisal of the situation and the subjective experience (Bullock & Russell, 1986). Among these cues, the representation of the causal bonds, that is a set

3 of causal events and of their behavioral consequences, has a remarkable significance, because they constitute the more explicative elements of the emotional experience (Want & Harris, 2001). Therefore, the concept of emotional context, considered as a complex and multidimensional representation of situational events, is relevant in facial expression processing. As Russell and Widen (2002) underlined, in everyday experience children use facial expressions in order to infer emotions. On the other hand, the facial cues are always located in an interactive context. A hypothesis that might account for autistic people difficulties is that the emotional and the social scripts are scarcely developed. Bullock and Russell (1986) proposed a model in which children acquire a system to represent and classify emotions which is based on a limited number of wide categories. The most important of them are the two dimensional axes of the hedonic value and the arousal level. This model was tested by some empirical studies which found that at the first time children interpret facial expressions in terms of pleasure displeasure (bipolar hedonic value) and intensity (arousal level). Only in a second time they use more articulated and wider conceptual categories (Widen & Russell, 2003). Through a progressive process of script generalization, a situated comprehension of emotions arises. This process is well illustrated by the use of adequate emotional linguistic labels. Indeed, emotional labels, initially few and more inclusive, progressively become more specific and articulated. Emotional labels constitute the final step of a developmental process that goes through a dimensional attribution (characterized by the presence of pleasure displeasure correlate) to a situational attribution (the script representation). Emotional labels belong to a more abstract level. To conclude, it is possible to maintain that the analysis of the processes underlying emotional expression decoding could provide important information regarding the conceptualization and lexicalization of facial expressions. This should be taken into consideration in studying the development of emotional decoding in autistic children, because these competences seem to be bound not only to cognitive but, above all, to social and communicative deficits, which have an influence on emotion conceptualization and lexicalization. 2. Aims and hypotheses M. Balconi, A. Carrera / Research in Developmental Disabilities 28 (2007) The present study aims to investigated the lexicalization and conceptualization of emotional facial cues in autistic and normal children. Specifically, it focuses emotional face representation in high-functioning school-aged autistic and normal children, through two different phases Phase 1 (facial expression of emotion) First of all, we investigated the system of verbalization and conceptualization of facial expression and its contribution to recognition and comprehension of emotions through two different levels of analysis. A first level concerned the ability to use a verbal label correctly (emotional label relevance); a second level concerned the wider lexical system used to describe emotional representation (emotional vocabulary) that can be considered as a marker of the acquisition of the semantic attribution of facial expression. (a) We suppose that autistic persons, if compared with normals, are generally less competent in the labeling and conceptualizing of an emotional representation. In particular, we suppose that they focus their representation on visual-perceptive attributes rather then on the

4 412 emotional correlates (such as causal boundaries etc.) and that their lexical production is less articulated and complex. (b) Furthermore, we expect that autistic people may have more difficulties to decode and understand secondary complex emotions (such as surprise and disgust) rather then primary basic emotions (happiness, sadness, anger, and fear). In particular, we focus our attention on the lexicalization and representation of the dimensional axis of hedonic value, that engenders the acquisition of a more complex conceptual representation (Balconi & Carrera, 2006). Thanks to this acquisition it can be produced the developmental process, that includes an initial competence in the discrimination of basic emotional categories (as happiness, sadness, fear) and a successive comprehension of more complex emotional categories (as surprise) Phase 2 (emotional script) We adopted a situational perspective to explain facial emotion comprehension. This perspective allows us to suppose that emotion decoding is the result of the elaboration of multiple emotional cues, among which facial patterns (facial expressions), behavioral correlates (the causal bonds between events), as well as specific contextual factors (eliciting emotional context). Therefore, we aim to analyze the whole of contextual cues that contribute to facial expression comprehension, thanks to the construction of an emotional script, namely the representation of an emotion expression in a defined context (the situated emotion ). The comparison between these two different types of condition (only a facial expression of an emotion; a facial expression within an emotional script) allows us to explore in detail the role of the eliciting context in the emotional representation and lexicalization. Also in this second phase, we analyze the same lexical categories of the first one (emotional label relevance and emotional vocabulary). (a) We suppose that script facilitates autistic subjects recognition, with respect to the categories of the emotional correlates, the eliciting causes and mental state of people. According to our hypothesis, a script will function as a facilitation cue to correctly interpret the whole emotional event. (b) This facilitation should be mainly more evident for the secondary emotions, as disgust and surprise, because in order to comprehend these emotions, subjects have to understand some contextual or external elements. 3. Method 3.1. Participants M. Balconi, A. Carrera / Research in Developmental Disabilities 28 (2007) The sample includes 26 normal and 7 autistic children. Ages varied from 6 to 11 (M = 8.75; S.D. = 0.78; range = ; 15 females and 11 males) for the normals. All these children attend the primary school at Milan or the province. None of them presented cognitive or linguistic deficits. The autistic sample is constituted by 7 subjects (age: M = 9.83; S.D. = 0.93; range = ; 2 girls and 5 boys) with a diagnosis of autism. With regard to cognitive competences, autistic participants presented a middle-high or high functioning cognitive profile (IQ: M = 87; range: ). The presence of other deficits on the perceptive or cognitive levels was excluded.

5 3.2. Materials M. Balconi, A. Carrera / Research in Developmental Disabilities 28 (2007) Facial stimuli (Phase 1) Fig. 1a shows the facial stimuli (cardboards black and white 10 cm 10 cm), which consist of an emotional face of a young boy showing six emotions (happiness, sadness, anger, fear, disgust and surprise). The stimulus material was selected by Ekman and Friesen database (1976). We have opted for a young actor aged similarly to the experimental subjects, in order to facilitate the identification process, which would make easier the recognition task Emotional scripts (Phase 2) The material consists of 6 pictures (coloured cardboards 10 cm 15 cm) with an emotional content (see Fig. 1b). Pictures illustrate contextualized situations eliciting the emotional correlates of happiness, sadness, anger, fear, surprise and disgust. In particular each picture presents a character (a girl or a boy) in an interactive context (with peers or adults). In addition, the presence of a clear emotional facial expression was considered a discriminant stimulus for the selection of the pictures. The pertinence of the emotional content for each emotional script, the homogeneity of the stimuli in terms of cognitive complexity and familiarity were tested in a Fig. 1. Materials: (a) facial expressions (Phase 1); (b) emotional scripts (Phase 2).

6 414 M. Balconi, A. Carrera / Research in Developmental Disabilities 28 (2007) pre-experimental phase (11 males and females; 6 11 years). The stimulus chosen at first to represent surprise was replaced, because it was frequently confused with other emotions (mainly with anger and sadness). Stimuli homogeneity and complexity were tested with a 5-points Likert scale (homogeneity F(5,10) = 1.24, p =.34; complexity: F(5,10) = 1.38, p =.27). In each phase, first time stimuli were presented simultaneously, in order to allow familiarization with the material. In a second assessment, they were presented one at time, in a random sequence, varying the order of the stimulus across the participants. Furthermore, to avoid a possible order effect between the experimental conditions, some subjects were submitted to face decoding condition (Phase 1) and successively to emotional script condition (Phase 2), whereas other subjects decoded the stimulus materials in an opposite sequence (firstly the facial expression and than the emotional script) Procedure Through a half-structured interview grid (Widen & Russell, 2004), the experimenter invited the child to observe the stimulus set and to describe the situation illustrated by the pictures ( What happened? ). Furthermore, it was made another focal question about the emotional feelings represented ( What made him/her feels like that? What is that facial expression? ). Subjects could freely describe the stimulus by means of the linguistic categories they preferred in order to explain what happened and to attribute an emotional state to the protagonist. Interviews were audio- and video-taped and scored verbatim. Three judges examined the verbal material encoded, in order to analyze specific conceptual categories relatively to: (a) correctness of the verbal labels (label relevance). Moreover, in order to represent exhaustively the verbal production of the subjects, further conceptual categories were considered, in addition to the label s relevance: an explicit reference to feelings (such as child feels angry ), and to his/her physical or mental states ( he feels bad ; he thinks an ugly thing ). (b) emotion conceptualization, that includes the following semantic categories: the eliciting causes (objects or persons, such as he has broken the toy ), the consequential effects ( now he smiles ) and the presence of a hedonic value ( positive/negative ; good/bad ) of the emotional content. For the first level of analysis (a), a correct answer included an explicit emotional label (such as happiness ) or synonymous terms ( joy ) (Widen & Russell, 2004). For the second level (b) the scoring was attributed on the basis of the logical relation of the verbal production with the semantic categories analyzed. An inter-judges reliability of 95% was considered significant (for a: 97.83%; for b: 97.13%). 4. Data analysis and results In order to analyze the effect of pathology (normal/autistic children), stimulus type (facial expression/script) and lexical category (presence/absence), a log linear hierarchical analysis with satured model was applied to each linguistic variable (see Areni, Ercolani, & Scalisi, 1994). A log linear analysis was conducted for each emotion. We report the statistical data relative to the main effects of pathology (2), task (2) and the presence/absence of the lexical category (2). Moreover, we report the interaction effects only when they are significant to the analysis (Table 1).

7 Table 1 Significant effects for log linear analysis fit model (in order to simplify, principal effect and interactions referred to the variable presence of the category are omitted) Disgust Happiness Fear Anger Surprise Sadness Label relevance Feelings Eliciting causes Physical or mental state Hedonic value = , p =.0011 x 2 (1, n = 33) = , p =.0006 Pathology, x 2 (1, n = 33) = 6.338, p =.0118 x 2 (1, n = 33) = , p =.0001 Pathology, x 2 (1, n = 33) = 7.987, p =.0047 Task, x2(1, n = 33) = , p <.001 x 2 (1, n = 33) = , p <.001 x 2 (1, n = 33) = , p <.001 Pathology, x 2 (1, n = 33) = , p =.0001 p <.001 = , p =.0006 x 2 (1, n = 33) = , p <.001 = , p <.001 = , p <.001 x 2 (1, n = 33) = , p =.0006 Pathology, x 2 (1, n = 33) = 5.948, p =.0147 x 2 (1, n = 33) = , p <.001 = , p <.001 x 2 (1, n = 33) = , p <.001 = , p =.0006 = , p <.001 x2(1, n = 33) = 4.458, p =.0347 Pathology, x 2 (1, n = 33) = 6.653, p =.0099 x 2 (1, n = 33) = , p <.001 = , p =.0002 Pathology, x 2 (1, n = 33) = 7.361, p =.0067 x 2 (1, n = 33) = , p =.0031 = , p =.0006 x 2 (1, n = 33) = , p =.0006 = , p <.001 x2(1, n = 33) = , p =.0010 Pathology, x 2 (1, n = 33) = 7.075, p =.0078 x 2 (1, n = 33) = , p =.0006 Pathology, x 2 (1, n = 33) = , p =.0006 = 4.738, p =.0295 x 2 (1, n = 33) = 6.062, p =.0138 = , p =.0006 = , p <.001 x 2 (1, n = 33) = , p =.0006 x 2 (1, n = 33) = 4.409, p =.0357 Pathology, x 2 (1, n = 33) = 5.549, p =.0185 x 2 (1, n = 33) = , p =.0002 = 5.131, p =.0235 x 2 (1, n = 33) = , p =.0001 Pathology, x 2 (1, n = 33) = , p <.001 x 2 (1, n = 33) = , p =.0006 = 4.464, p =.0346 x 2 (1, n = 33) = , p =.0006 = , p <.001 x2(1, n = 33) = , p =.0006 = , p <.001 x 2 (1, n = 33) = , p =.0009 M. Balconi, A. Carrera / Research in Developmental Disabilities 28 (2007)

8 416 M. Balconi, A. Carrera / Research in Developmental Disabilities 28 (2007) Table 2 Percentage values for each verbal category (script and face) Disgust Happiness Fear Anger Surprise Sadness Script Face Script Face Script Face Script Face Script Face Script Face Label relevance Normal Autistic Feelings Normal Autistic Eliciting causes Normal Autistic State Normal Autistic Hedonic value Normal Autistic In Table 2 we report the percentage values for each linguistic category Happiness In both recognition tasks (emotional face and script), happiness was largely recognized by the subjects, normals and autistics. Also the emotional correlate was referred by almost all the sample. Above all it was used by normals in the script decoding task. Also, causal antecedents, especially in the case of normal children, were frequently produced in the script decoding. They were rarely reported in the face description. In particular, in this task no autistic child cites causal bonds. As regard to physical or mental states, in the script description it was often represented by autistic children, while on the contrary in face recognition only normals frequently make reference to them. Finally, taking into account the variable of pathology, it is interesting to note a significant difference in the representation of the hedonic value: many normal children reported this semantic category, which instead was ignored by all autistic participants Fear Fear was correctly identified by the whole sample. Emotional correlate was largely used in both conditions. Indeed, all normals make use of this category in the script recognition, while in the face decoding all autistic children explicited the emotional label. In general, the causal antecedent for fear was scarcely reported, even if it is noticeable that in the script recognition almost all autistics described the causes of the emotion. Physical or mental states were broadly used by normals to describe the emotional script, while it was rarely present in autistics report. In the face recognition task, both normal and autistic children made few references to the state. Finally, hedonic value was sometimes reported by normals, while no autistic child mentioned this category.

9 M. Balconi, A. Carrera / Research in Developmental Disabilities 28 (2007) Anger Anger was largely recognized, especially by autistic children. Also, the emotional correlates were frequently expressed, above all by normal children. Anger causal antecedents were largely present in the script decoding task, while they were less frequently used to describe facial expression. This trend was more evident for autistics: all the subjects employed this conceptual category to characterize script, while none of them knew the causes in the case of face recognition. The state was rarely reported in both tasks. In particular, no autistic person described the physical or mental state related to facial expression. Finally, the hedonic value was never reported by autistics, but it was frequently named by normals, mainly during the script description Surprise On the contrary, surprise was scarcely labelled, especially as regard to the script comprehension. In this task normals showed a better performance in recognizing the emotions, while in the face decoding task autistic people revealed an improved ability over normals (more than 50%). Emotional correlates were often expressed, above all in the script representation. Both normals and autistics frequently made use of some causal antecedents in describing the script, while they rarely cited the causes for the facial expression. This trend was particularly accentuated in the case of autistic subjects. Finally, almost all of the normal subjects described the hedonic value of emotions, while no autistic child used this category Sadness In both of the recognition tasks, this emotion was largely labeled by almost all of the subjects, normal and autistic. The emotional correlates for sadness were expressed by the subjects, mainly by the autistics. In general, causal bonds were largely described as regard to the script but not to the face. Particularly, all autistic children described the causes for the script, while none of them used causal cues for facial expression. Moreover, state category was scarcely represented, especially in the facial expression recognition, even if almost all autistic children used it in describing the script. Finally, hedonic value was absent in autistic children verbalization, while it was frequently referred to by normals, mainly during the script decoding Disgust Disgust was scarcely recognized by the participants, with a worse performance for facial expression than for script, and this difficulty was more accentuated for autistics. On the contrary, emotional correlates were frequently reported, above all by normals. Emotional script was more often described in terms of causal antecedents. In particular, normals made use more frequently of the causal cues, which were never mentioned by autistics. Conversely, only autistic children made reference to the causes that determined the emotional script. Finally, hedonic value was never used by autistics, while it was widely reported by normals in both of the tasks. 5. Discussion The present study produced three major results: first, there was a clear differentiation in children s conceptualization and lexicalization as a function of different emotions; besides, the

10 418 M. Balconi, A. Carrera / Research in Developmental Disabilities 28 (2007) presence of two different types of task a facial expression decoding and a script comprehension activated different representational levels; finally, it was possible to identify some differences in emotion representation between normal and autistic subjects Characterization of different emotions The emotions of happiness, fear, anger and, almost in part, sadness are well recognized and well labelled. As regard to fear, a recent study by Castelli (2005) revealed high rates of recognition, that the researcher attributes to the central adaptive function of this emotion. Indeed, the emotion of fear, both on an ontogenetic and a phylogenetic level, has a main role for the individual safeguard. It is exploited as a cue in order to detect unfavorable environmental conditions (Balconi, 2004; Balconi & Pozzoli, 2003a). Accordingly to the functional model (Frijda, 1994; Lazarus, 1999), the emotional expressions represent a response to a particular event, significant in terms of costs and benefits for people. In these terms, the expression of fear represents the perception of a threat for the personal safeguard and, therefore, it requires a greater investment of attentional resources. On the contrary, surprise and disgust appear to be very difficult to recognize. Only a few subjects were able to name them with an adequate verbal label, and these emotions were frequently confused with other emotions. In particular, surprise was often confused with fear, while disgust lacks of an explicit recognition by the subjects ( I don t know ). As hypothesized by the dimensional approach to emotion (Balconi & Pozzoli, 2003b; Ellsworth & Scherer, 2003), the representation of the emotional domain was based on a conceptual space defined by two exes, arousal and hedonic value. In particular, the emotions with a high arousal level and a negative value are better understood, if compared with other emotions (see Fig. 2). Moreover, the results of the present research underline that causal category was a main feature in children representation and lexicalization, because it contributes to the logical bonds and antecedent-consequent relations (Ellsworth & Scherer, 2003). In particular causal correlates that are directly linked to the interactive dynamics between persons are more frequently described by children (see Fig. 3). Fig. 2. Label relevance (rate values): N, normal, A, autistic.

11 M. Balconi, A. Carrera / Research in Developmental Disabilities 28 (2007) Fig. 3. Explicitation of the causal correlate (rate values). This trend was of primary importance in our study, particularly for happiness, anger, surprise and sadness. Indeed, causal factors were more frequently used to represent script, rather then facial expression. A particular case was represented by fear. Its causal antecedents were often described by the subjects, and they were referred by autistic children in a significant manner, above all in the script description. It is interesting, because fear is an emotion which can be simply understood in terms of external causes and it does not require a representation of mental states, as in the case of sadness. In fact, the latter is often described taking into account its causal antecedents, but in a different way. Only normal children present a wide representation of sadness, considering it was derived from a negative intra-subjective condition, while autistic children make reference to an external cause, that is to an inter-subjective condition ( it is guilty of... ). It was likely that autistic children need of an external justification, while normal ones represent sadness as an internal construct, in absence of an objective eliciting motive. If it was so, normal children showed a greater mastery of psychological dimension related to sadness than autistics. Surprise shows a representation that was quite different. It was rarely recognized (only 50% of the sample correctly recognizes this emotion), but in this case both of the emotional correlates and the causal relations were well represented. Moreover, in the case of surprise, the negative hedonic value was correctly reported. In our study, the dichotomy pleasure/displeasure was spontaneously and frequently verbalized by the subjects, confirming a significant role in emotion representation, as indicated by previous researches (Balconi, 2004; Balconi & Pozzoli, 2003a; Ellsworth & Scherer, 2003). In fact, not only the hedonic category was systematically verbalized, but it was correctly identified in terms of negativity or positivity. Moreover, it can be considered a predictive cue of the ability to classify and differentiate emotional correlates. Indeed, it was correctly verbalized only when the child was able to attribute an adequate label to the emotion, while the child cannot conceptualize the emotion, the hedonic value seems to be ambiguous and not very well developed. This was the case for the emotions surprise and disgust. They were scarcely characterized in terms of a specific lexical category, as well as of a hedonic attribution, especially for autistic children. Generally, the hedonic category representation is differentiated on the basis of the subject profile. Indeed, it was consistently used by normals to describe emotions, while it was infrequently present in the verbalizations of autistic people (see Fig. 4).

12 420 M. Balconi, A. Carrera / Research in Developmental Disabilities 28 (2007) Fig. 4. Explicitation of the hedonic value (rates values) Pathology and task effects As regard to the pathology, lexicalization and conceptualization of the emotional universe appear to be well defined both for normal and autistic participants, even if with some relevant differences. Indeed, emotions were frequently and correctly recognized by both sub-samples, even if the representation was not so well developed for the secondary emotions, that was for the emotion of surprise and disgust. The similarities between normal and autistic children suggest an analogue process of emotion acquisition, which proceeds from the acquisition of a primary level to a secondary level. The situated emotion condition (script decoding) introduces another main explicative factor, regarding emotional representation. Indeed, the presence of a specific context does not affect the correctness of the emotional label attribution, but it produces a greater difference in a wider lexical competence, which includes the representation of causal components, of the emotional correlates and physical and mental states. Verbalizations elicited in describing facial expressions were shorter and poorer with respect to content, if compared with that elicited by the emotional script, which was conceptually more articulated, since it includes well-planned references to the conceptual category focalized. It was possible to state that the situational component constitutes a facilitation cue, because it allowed the subjects to activate a more complex conceptual representation, which takes into account the context in which the emotional event happens, the emotional causes, the sequence of actions and their consequences (Bullock & Russell, 1986). It was noticeable, however, that the script enables a wider and a more complete representation, especially in the case of the autistic children, who maximally have a benefit from the situated condition. The presence of the interactional features that characterize the emotional experience constitutes a facilitation element for emotion comprehension, also producing a better description in the emotion labeling. In particular, above all, for autistic participants, emotional script improves the subjects performances with respect of the emotions of fear, anger, and surprise. Conversely, face decoding process appears to be more difficult for the whole sample. The representation was not always adequate in terms of lexical relevance with respect to the whole comprehension of the stimulus, as in the case of representation of the emotional correlates and the hedonic value.

13 M. Balconi, A. Carrera / Research in Developmental Disabilities 28 (2007) To conclude, even if our study does not allow us to state which of the two representational modalities (facial pattern comprehension or script decoding) precedes the other, it was possible to observe that the situational correlates provide a facilitation cue for the representation of emotion. Finally, it was useful to state that our study investigated verbal production using a freelabeling method, that allowed to avoid possible artifact in child conceptualization. On the other hand, this methodology limited the exploration to semantic categories spontaneously produced by the subjects. Future research could aim to integrate this method, for example adding some specific conceptual categories to be explored. References Areni, A., Ercolani, A. P., & Scalisi, T. G. (1994). Introduzione all uso della statistica in psicologia (Introduction to the use of statistics in psychology). Milano: LED. Balconi, M. (2004). Neuropsicologia delle emozioni (Neuropsychology of emotion). Roma: Carocci. Balconi, M., & Carrera, A. (2006). Lessicalizzazione e concettualizzazione nella comprensione della mimica facciale. Analisi mediante script emotive [Lexicalization and conceptualization in facial mimic comprehension. An emotional script analysis]. Età Evolutiva, 83, Balconi, M., & Lucchiari, C. (2005). Consciousness, emotion and face: An event-related potentials (ERP) study. In R. D. Ellis, & N. Newton (Eds.), Consciousness & emotion. Agency, conscious choice, and selective perception (pp ). Amsterdam/Philadelphia: John Benjamins Publishing Company. Balconi, M., & Lucchiari, C. (2006). EEG correlates (event-related desynchronization) of emotional face elaboration: A temporal analysis. Neuroscience Letters, 392, Balconi, M., & Pozzoli, U. (2003a). Face-selective processing and the effect of pleasant and unpleasant emotional expressions of ERP correlates. International Journal of Psychophysiology, 49, Balconi, M., & Pozzoli, U. (2003b). Semantic attribution and facial expression of emotion. Consciousness & Emotion, 2, Baron-Cohen, S., Spitz, A., & Cross, P. (1993). Do children with autism recognize surprise? A research note. Cognition & Emotion, 7, Bormann-Kischkel, C., Amorosa, H., & von Benda, U. (1993). Is there an association between emotional feelings and emotional signs in autism? Acta Paedopsychiatrica, 56, 1 9. Bormann-Kischkel, C., Vilsmeier, M., & Baude, B. (1995). The development of emotional concepts in autism. Journal of Child Psychology & Psychiatry, 36, Bullock, M., & Russell, J. A. (1986). Concepts of emotion in developmental psychology. In Izard, C. E., & Read, P. B. Eds. Measuring emotions in infants and children. Vol. 2 (pp ). Cambridge: Cambridge University Press. Capps, L., Yirmiya, N., & Sigman, M. (1992). Understanding of simple and complex emotions in non-retarded children with autism. Journal of Child Psychology & Psychiatry, 33, Castelli, F. (2005). Understanding emotions from standardized facial expressions in autism and normal development. Autism. The International Journal of Research and Practice, 9, Celani, G., Battacchi, M. W., & Arcidiacono, L. (1999). The understanding of the emotional meaning of facial expressions in people with autism. Journal of Autism and Developmental Disorders, 29, Ekman, P., & Friesen, W. V. (1976). Pictures of facial affects. Palo Alto: Consulting Psychologist Press. Ellsworth, P. C., & Scherer, K. R. (2003). Appraisal processes in emotion. In R. J. Davidson, K. R. Scherer, & H. H. Goldsmith (Eds.), Handbook of affective sciences, Series in affective science (pp ). London: Oxford University Press. Fein, D., Lucci, D., Braverman, M., & Waterhouse, L. (1992). Comprehension of affect in context in children with Pervasive Developmental Disorders. Journal of Child Psychology & Psychiatry, 33, Frijda, N. H. (1994). Varieties of affect: Emotions and episodes, moods and sentiments. In P. Ekman, & R. Davidson (Eds.), The nature of emotion: Fundamental questions (pp ). Oxford: Oxford University Press. Gepner, B., de Gelder, B., & de Schonen, S. (1996). Face processing in autistics: Evidence for a generalized deficit? Child Neuropsychology, 2, Hillier, A., & Allinson, L. (2002). Beyond expectations. Autism, understanding embarrassment, and the relationship with theory of mind. Autism. The International Journal of Research and Practice, 6, Kanner, L. (1943). Autistic disturbances of affective contact. Nervous Child, 2,

14 422 M. Balconi, A. Carrera / Research in Developmental Disabilities 28 (2007) Lazarus, R. S. (1999). The cognition-emotion debate: A bit of history. In T. Dalgleish, & M. J. Power (Eds.), Handbook of cognition and emotion (pp. 3 19). Chichester: John Wiley. Rojahn, J., Lederer, M., & Tassé, M. J. (1995). Facial emotion recognition by persons with mental retardation: A review of the experimental literature. Research in Developmental Disabilities, 16, Russell, J. A., & Widen, S. C. (2002). Words versus faces in evoking preschool children s knowledge of the causes of emotions. International Journal of Behavioral Development, 26, Want, S. C., & Harris, P. L. (2001). Learning from other people s mistakes: Causal understanding in learning to use a tool. Child Development, 72, Widen, S. C., & Russell, J. A. (2003). A closer look at preschoolers freely produced labels for facial expressions. Developmental Psychology, 39, Widen, S. C., & Russell, J. A. (2004). The relative power of an emotion s facial expression, label and behavioral consequence to evoke preschoolers knowledge of its causes. Cognitive Development, 19,