قدرت نسبی احساسات و صورت بیان، برچسب و پیامدهای رفتاری به برانگیختن دانش پیش دبستانی با توجه به علت آن

Abstract Lay people and scientists alike assume that, especially for young children, facial expressions are a strong cue to another’s emotion. We report a study in which children (N=120; 3–4 years) described events that would cause basic emotions (surprise, fear, anger, disgust, sadness) presented as its facial expression, as its label, or as its behavioral consequence. For no emotion was the facial expression the strongest cue. Performance for fear and disgust was more accurate given its label or its behavioral consequence than given its facial expression; performance for anger was more accurate given the consequence. For 3s, behavioral consequences were the strongest cues to emotion; for 4s, labels were.

3. Conclusion In this study, 3- and 4-year-olds were asked to imagine what might have caused another child’s emotion. The task per se was within their abilities: 100% generated a plausible causal story for happiness. For the other five emotions, although the children generated stories 89% of the time, the stories were correct only about half the time (43% by best guess, 63% by plausibility). What was interesting was that their performance describing a cause varied with the cue to the other’s emotion. Showing children the hypothesized universal, biologically based facial expression for a basic emotion was overall the least good way to specify that emotion. Our results thus challenge the assumption that prototypical facial expressions are especially potent cues to emotion for 3- and 4-year-olds. At this age, labels or behavioral consequences were both stronger cues overall. Although the facial expression mode was relatively weak in this study, we are not willing to abandon the notion that facial expressions have a role to play in the development of children’s understanding of emotion, although perhaps a more limited role than sometimes envisioned. Especially for children younger than 3 years, and especially for those emotion concepts acquired first (happiness, sadness, anger), there is ample evidence to support a role for facial expressions. Nevertheless, the question must be raised concerning exactly how these young children interpret those facial expressions and to what use they put their interpretations. We suggest that before their third birthday, children interpret facial expressions in terms of broad categories, based largely on pleasantness-unpleasantness and, secondarily, arousal (Widen & Russell, 2003a). Facial expressions so interpreted provide the toddler with a broad but relatively undifferentiated structure for emotion. We thus replicated the Label Superiority Effect: Performance was higher overall given an emotion’s label than given its facial expression. The Label Superiority Effect has thus now been found with three different methods (free labeling, sorting, and specifying a cause), suggesting that labels are powerful cues to emotion in different contexts. In the current study, the Label Superiority Effect was significant overall, and strongest for fear and disgust; but weak for sadness; and absent for anger and surprise. These differences across emotions also replicate prior findings (Camras & Allison, 1985, Russell, 1990 and Russell & Widen, 2002b). What was especially interesting in the current study was that the effect of labels was significant for 3s and even stronger for 4s. Indeed, improvement with age was limited to the label mode of presentation. Labels might play a more important role in the development of emotion concepts, but mainly when 4-year-olds are acquiring the concepts of fear and disgust. In contrast, at a younger age, and for concepts of happiness, sadness, and anger, which are acquired earlier (Widen & Russell, 2003a and Widen & Russell, 2003b), labels were not as powerful—suggesting that labels might play a lesser role. In later development, children acquire concepts for emotions with no strong link to any unique facial expression—such as love, hate, pride, guilt, shame, and so on. Perhaps in these cases, labels will prove to be especially important. We also found a Behavior Superiority Effect: Performance was higher given an emotion’s behavioral consequence than given its facial expression (Fig. 1). This effect was strongest for anger, but present for fear and disgust as well. It was weak for sadness; and reversed for surprise (both label and facial expression were better than the specific consequence we offered children). Our study may have underestimated the power of behavioral consequences. The emotion labels (sad, scared, etc.) that we presented to the children are likely close to the best instances possible for the label mode of presentation. Similarly, the facial expressions we presented are likely close to the best instances possible, being clear (perhaps even exaggerated) prototypical signals based on Ekman and Friesen’s (1978) specifications. The specific labels and facial expressions we used were chosen on the basis of extensive prior work on what is most likely to be understood by this age group. Furthermore, labels presented orally and facial expressions presented visually likely mirror the modality under which children encounter labels and facial expressions in their everyday experience. In contrast, the behavioral consequences we used, although our best effort, were a first effort and might well be subject to improvement through trial-and-error. In addition, the behavioral consequences were presented orally, a modality that although not unprecedented might be less common and less salient for children than witnessing actual behaviors. It may therefore be possible to strengthen the Behavior Superiority Effect by fine-tuning the behavioral consequences used for each emotion and by presenting them pictorially. Despite these limitations, behavioral consequences were the best of the three modes of presentation overall for 3-year-olds. Behavioral consequences were also especially powerful for the emotion of anger, and anger is one of the earlier emotion concepts acquired, preceded only by happiness (used here as a screening trial) and, in some children, sadness (Widen & Russell, 2003a). The present results offer an intriguing possibility: Especially for 3-year-olds, and especially for anger, an emotion’s behavioral consequence might be the most salient cue to forming the concept of that emotion. The present results are the first demonstration of which we are aware that behavioral consequences can be a powerful cue to the emotion of another for a preschooler. Should this effect prove replicable in different contexts and with different methods, it could provide another insight into the younger preschooler’s development of emotion concepts. Taken together, our results suggest a much more interesting developmental story. These results are more in keeping with a script theory than with notions of a single origin for emotion categories. According to a script theory, the features of an emotion (cause, physiology, behavior, facial expression, subjective experience, label, etc.) tend to be correlated; with time and experience, children learn these associations and develop scripts of prototypical emotional events (Bullock & Russell, 1986 and Fehr & Russell, 1984). Let us speculate that scripts for happiness and sadness might begin with facial expressions—smiling and crying—that become linked to their causes. For anger, the script might begin instead with hostile behavior being linked to its cause. For fear and disgust, the script might begin with someone using the labels scared and disgusted, which becomes linked to their causes. In each case, once the first link is formed the child can then add additional information. Because different cues initiate the process for different emotions, different cues are more powerful for different emotions. Because concepts for different emotions are acquired at different ages, different cues are more powerful at different ages. The present study was limited to the particular method used, and replication with other methods is required before these hypotheses can be given much credence. For example, with our method, the child told of a causal event for a given cue. It is possible that the child simply associated the causal event directly with the cue, bypassing the emotion altogether. Thus, the child who responded, “someone knocked Danny’s tower down” might have imagined this as the cause of the frown (the facial expression of anger), of the use of the word angry, or of Danny’s reaction of hitting and yelling—all without really understanding that Danny was angry. This possibility can be approached as a methodological problem, and from the network of other research on 3- and 4-year-old children, we are confident that they can be shown to do more than simply associate the causal event with the cue. But there is also a conceptual issue involved. On the script hypothesis just described, a child begins with a pair of events (such as a frustrating event and frowning). From this seed, the script for anger is built by adding more elements. The adult knows anger’s subjective feelings and thoughts, its range of behaviors, its physiological manifestations, its social norms, its many contexts, its various means of expression, and so on. More generally, concepts for specific emotions consist of multiple components, which are likely added to the child’s knowledge base one at a time. The conceptual question is this: At what point, in this progress from a pair, then a triplet, on to finally a multicomponent adult script does the child “have” the concept of anger? There may be no empirical answer to this question, but rather a fuzzy boundary from not having to having the concept. Another methodological and conceptual limitation of our study is its focus on children’s understanding of the emotions of others. Their understanding of their own emotions was therefore invisible. For example, when witnessing his dad getting angry, Dad’s frown may be salient. In contrast, when Chris is angry himself, his own frown (especially its visual appearance) is less salient, perhaps inaccessible. The relative weakness found here for facial expressions is consistent with the possibility that self-experience plays an important role in the development of emotion concepts. Indeed, half of children’s spontaneous references to emotions in conversation are to their own emotions (Wellman, Harris, Banerjee, & Sinclair, 1995). This important but difficult-to-study component must also be considered.