1 Preschoolers Interpretation of Facial Expressions of Emotion Sherri C. Widen & James A. Russell Boston College Presented at Conference on Human Development, Charlotte NC, April, 2002 Abstract Asked to label the emotion expressed by a prototypical facial expression of a basic emotion, children (N = 160; 2 to 5 years) make many errors and improve only gradually with age. Contrary to the usual interpretation of this standard finding, we found that errors are not artifacts of poor stimuli or the difficulty of a production task: errors remained even with excellent examples of prototypical expressions and with children who can perform a production task (labeling animals). Errors reveal rather than conceal children s way of interpreting facial expressions. Children s use of emotion labels increases in a systematic order:, sad, and angry emerge early, are more accessible, and are applied broadly but systematically. Scared, surprised, and disgusted emerge later, are less accessible, and are applied narrowly. Accessibility of the later three labels increases with age. Introduction An experimenter presented children (N = 160; 2 to 5 years of age) with six photographs of facial expressions of emotion. She asked, How does Sally feel in this picture? As had other researchers who used this free labeling method (Izard, 1971; Harrigan, 1984; Wiggers & van Lieshout, 1985; Markham & Adams, 1992), we found that the proportion of correct responses was modest and improved only gradually with age. Free labeling is interesting because it is the method that seems to come closest to tapping a child s spontaneous specification of the emotion seen in a face. A key question must be answered, however, before such data can be properly interpreted: Why do children make errors on this task? One obvious possibility is that children make errors when they do not know the meaning of the face, at least in terms of what would be considered correct by the researchers. On this interpretation, children mislabel the sad
2 face because they do not interpret a person displaying it as sad. If this is the case, children s errors reveal something of their actual understanding of a facial expression. Children s free labeling responses are much richer than implied by analyses of accuracy. On this task, children are not simply silent when they are not producing the correct label. Rather, they frequently provide incorrect labels. In the current study, we were interested in all the responses children made, regardless of correctness. On our view, free labeling data provide a window to children s current understanding of emotion and interpretation of facial expressions. Thus, we were interested in two questions: What specific labels did children use? How many different labels did each child use? The answers to these two questions led to a data-driven model of the acquisition emotion categories and labels. Method Participants. Participants were 160 children (80 girls, 80 boys). There were 40 children (20 girls and 20 boys) in each of four age groups: 2s (48 to 35 months; mean = 30 months, s.d. = 3.7), 3s (36 to 47 months; mean = 42 months, s.d. = 3.6), 4s (48 to 59 months; mean = 53 months, s.d. = 3.3), and 5s (60 to 71 months; mean = 65 months, s.d. = 3.1). Materials Photographs of Animals. The animal pictures were six color photographs, one each of a cat, dog, horse, cow, turkey, and goose, included in the book Who are you? Country friends (1992). Photographs of facial expressions. Each of seven photographs (one each for happiness, sadness, anger, fear, disgust, surprise, neutral) was of the same 13-year-old girl (Figure 1). The photographs were provided by Dr. Linda Camras. Camras, Grow, and Ribordy (1983) described the development of the photographs, their coding according to Ekman and Friesen s (1978) Facial Action Coding System, and their use in a study on recognition of emotional expressions.
3 Figure 1. The faces used in the current study. Procedure The experimenter spent the first visit getting to know each child. On a subsequent visit, the experimenter invited an individual child to look at the special books she had brought with her. The pictures of animals were presented in their original book format, and the photographs of facial expressions were placed one per page in a photo-album. Each child participated in two sets of trials. The first set of trials concerned animals; the second set, facial expressions. The animal trials served as a practice session and a comparison task. The experimenter said, This is my special book about animals. [Opening the book and pointing to the first page.] Do you know what kind of animal this is? Responses were not corrected and were mildly praised equally. Six animals were shown, always in the same order. The experimenter ended by saying, That was really fun. Do you want to look at my second book? For the second set of trials, the order of the emotional facial expressions was varied randomly, except that the first trial was always the smile, because piloting indicated that even young children frequently labeled this expression as happy.
4 Results The results of the current study replicated prior findings on children s performance on free labeling facial expressions: Performance was modest and improved only gradually with age. Overall, the children produced the correct label on 42.9% of the 960 trials. Proportion correct varied both with face, F(5, 780) = , p <.001, and with age, F(3, 156) = 50.55, p <.001. In addition, the face x age interaction was significant, F(5, 780) = 8.28, p <.001 (Figure 2). Ruling Out Method Artifacts Are these results due to method artifacts? One method artifact is the quality of the facial expressions. The facial expressions used in the current study were FACS coded. Each face had the AUs of a prototypical facial expression of a basic emotion. Thus, they were clear examples of each of the target emotion. The results of the animal labeling task suggest that this modest performance was not due to the difficulty of the production task per se. On the animal labeling trial, the children were correct on 73.8% of the 930 trials. For the four mammals, the children did very well: Even the 2s produced a high proportion of correct labels (85% to 97%). Every child named at least two mammals correctly. These results were clear evidence that even the youngest 2-year-old understood the instructions and was willing and able to produce a label. A third method artifact is label accessibility. Specifically, if children didn t know the target label, or if was not accessible, then the children obviously could not correctly label the faces. However, observational studies and parental reports of the emotion labels that children know indicates that by their second birthday children know and use happy, sad, angry, and scared; surprised and disgusted emerge before their fourth birthday (e.g., Ridgeway, Waters, & Kuczaj, 1985; Wellman, Harris, Banerjee, & Sinclair, 1995). In addition, prior to free labeling, Widen and Russell (2002) used Active Priming (during which each child responded to questions intended to elicit each of the six target emotion terms). Each of the children (N = 80; 3 to 4 years of age) produced all six emotion labels, thereby indicating that the terms were accessible. As a result of the priming, children s performance for happiness, sadness, and anger was at ceiling. But their performance for
5 fear, surprise, and disgust was low (range:.13 to.68), and improved only gradually with age. Thus, even when each of the target emotion labels were accessible, they were not always associated with the facial expressions for those emotions. Children s Use of Emotion Labels The number of different emotion labels used increased over the span of 2 to 5 years of age (Figure 2). The frequency with which children used specific labels was: angry (220), happy (196), sad (117) scared (50), surprised (48), and disgusted (12). The same pattern occurred for both correct uses and incorrect uses, suggesting that the pattern is not a result of the faces per se, but rather of children s interpretation of those faces. Thus, young children favor some categories (happiness, sadness, anger) and tend to ignore the rest happy sad scared surprised angry disgusted Age (years) Figure 2. Proportion of children who correctly labeled each face. By 3 years of age, children were accurate at labeling the happiness and anger expressions. By 4 years of age, most children accurately labeled the sadness face. And the majority of 5-year-olds correctly labeled the surprise face. Improvement for the fear face was more gradual, and for the disgust face there was little improvement with age.
6 Specific labels emerged systematically during development (Figure 3). All children were sorted, irrespective of age, by the number of labels they used. We found that if a child used only one label, that label was most likely to be happy. For two labels, Number of Emotion Labels (mean age in months) 0 (30 mo.) 1 (39 mo.) 2 (40 mo.) 3 (50 mo.) 4 (56 mo.) 5 (62 mo.) 6 (56 mo.) No label     Surprised  Scared  Surprised Scared  Surprised Scared Disgusted  Figure 3. The order in which emotion labels emerged there were two possible paths. Some children added angry, others added sad. For those who used three labels, the two paths merged and all the children used happy, angry, and sad. The next step again allowed two paths: some children added surprised and some scared. For five labels, the two paths again merged. In the last step, disgust was added. Figure 3 represents a simple model of the emergence of emotion labels; 81.3% of children fit the pattern shown a proportion significantly greater (p <.001) than the 23.3% expected by chance.
7 Conclusion Asking children to examine a good photograph of a prototypical facial expression and to say what the person shown is feeling turns out to be revealing task. It is not, we believe, dominated by simple method artifacts (task difficulty, quality of stimuli). When children s responses are simply categorized as right or wrong, the main conclusion from this method has been that the task is too difficult for children. But examining the errors proved quite interesting. Let us conclude with some thoughts about the development of children s responses to this task. The most basic level of performance we found (with children past their second birthday) was no association of labels with facial expressions: eighteen children in Study 2, seventeen of whom were 2-year-olds (mean age: 2;6), used no labels. Lack of a response is difficult to interpret, but recall that all these same children had just labeled three mammals (and were correct on at least two of them). At 2 years of age, emotion label availability is questionable (in Study 3, availability was established for 3- and 4- year-olds; but 2-year-olds were not included in that sample), although other aspects of the data caution against assuming unavailability necessarily accounts for their lack of labeling. Other method artifact explanations (poor stimuli or presentation, difficulty of production task) play a minor role at best. Investigation of this level, mainly in 2-yearolds, is a high priority for future research. The next level is the use of one label, most likely happy (mean age: 3;3 years). For children at this level, the meaning they associate with happy is difficult to characterize. Their responses to animals suggest that they might use broad categories. Perhaps, happy at this stage means something like emotional. This possibility, too, merits further research. The next level is the use of two labels, one positive (happy) and one negative (either sad or angry). This third level (mean age: 3;4 years) quickly follows the second. In this level, happy means positive and angry (or sad) means negative. The next level is three labels, with negative emotions divided into angry and sad. This level is achieved on average about ten months after level 3 (mean age = 4;2 years). Each of these three labels was applied broadly by both younger and older children. We suggest that happy continues to mean positive, whereas negative emotions are divided
8 into those with high arousal (angry) and those with low arousal (sad). Our results on the breadth of these three labels supports this interpretation. In the next three levels, the words surprised, scared, and disgusted were added (often in that order) to the child s repertoire for interpreting facial expressions. Unlike happy, sad, and angry, however, these last three labels are applied much more narrowly from the beginning. Even when in the child s vocabulary, these labels remain relatively low in accessibility. The initial three labels (happy, sad, angry) continue to be used for the expressions of fear, surprise, and disgust, even as these new labels are added. This result hints at an explanation for the seemingly slow pace at which children improve in correct free labeling: they already have a way of interpreting faces that presumably serves reasonably well. Rather than replacing no knowledge with knowledge, they are fine-tuning their way of interpreting faces. The question therefore arises of just what propels this change. References Camras, L. A., Grow, J., & Ribordy, S. (1983). Recognition of emotional expression by abused children. Journal of Clinical Child Psychology, 12, Ekman, P., & Friesen, W. V. (1978). Facial action coding systems. Palo Alto, CA: Consulting Psychologists Press. Harrigan, J. A. (1984). The effect of task order on children s identification of facial expressions. Motivation and Emotion, 8, Izard, C. E. (1971). The face of emotion. New York: Appleton Century Crofts Markham, R. & Adams, K. (1992). The effect of type of task on children s identification of facial expressions. Journal of Nonverbal Behavior, 16, Ridgeway, D., Waters, E., & Kuczaj II, S. A. (1985). Acquisition of emotion-descriptive language: Receptive and productive vocabulary norms for ages 18 months to 6 years. Developmental Psychology, 21, Wellman, H. M., Harris, P. L., Banerjee, M., & Sinclair, A. (1995). Early understanding of emotion: Evidence from natural language. Cognition and Emotion, 9, Widen, S. C. & Russell J. A. (2001). A Closer Look at Preschoolers Freely Produced Labels for Facial Expressions. Manuscript submitted. Wiggers, M. & van Lieshout, C. F. M. (1985). Development of recognition of emotions: Children s reliance on situational and facial expressive cues. Developmental Psychology, 21,