An Audience Effect on Smile Production in 10-month-old Infants

This report presents evidence that smile production in lO-month-old infants is affected by the presence or absence of an audience for the facial display. The audience effect does not appear to be mediated by emotion. The evidence indicates that the production of facial expressions is at least partly independent of emotion and partly dependent on a social-communicative context from a very early age. Smiles are interpreted as outward signs of positive affect. Yet smiling by adults depends on social contact as well as on the smiler's happiness (Fridlund, 1989. 1990; Kraut & Johnston, 1979). For example, bowlers who score a strike or a spare usually do not begin to smile when they see the pins fall. Instead, they smile after turning to face their companions. Solitary bowlers who score well usually do not smile al all (Kraut & Johnston. 1979). Recent studies have reported such "audience effects" on the frequencies of adults" empathic pain grimaces (Bavelas et a!., 1986) and facial reactions to odors (Gilbert, Fridlund, & Sabini, 1987), as well as on adult smiling (Fridiund, 1989; Kraut & Johnston, 1979). In each case, emotion was expressed in the face more often when other people were watching than when there was no social audience for the facial expression. In the present study, we asked whether human infants also show audience effects on their smiling behavior. We made videotapes of 10-month-oId babies in an infant analogue to a bowling alley and looked for effects of the availability of a social audience on the frequencies and directions of the infants' smiles. This research will add to our knowledge of the developmental origins Correspondence and reprint requests should be addressed to Susan S. Jones, Psychology 353, Indiana University, Bloomington, Indiana 47405. of social control over facial expressions. However, the principal theoretical reasons for this work concern a more general issue in the study of emotion and facial behavior. What is the relation between the emotional-expressive and social-communicative roles of facial expressions? Currently, two major schools of thought can be distinguished. One widely held view regards certain universal facial expressions as components of basic human emotions (Ekman, 1971, 1977; Izard, 1971; Izard & Malatesta, 1987; Tomkins, 1962). By this view, emotional experience automatically gives rise to prewired patterns of expressive movement. These prewired facial movements go forward unless they are actively inhibited or modified by "top-down" processes. By this view, then, smiles and other standard expressions are primarily emotional and secondarily social. Social control is overlaid on the automatic emotional signal in development, with the acquisition of the social motivations for deception and the ability to voluntarily control the facial musculature (Ekman, 1985; Ekman & Friesen, 1971; Malatesta, 1985; Saarni, 1978; Zivin, 1982). A second, emerging view focuses on the social-communicative and socialregulatory role of facial movements (Fridlund. 1990). By this view, human facial expressions are first and foremost evolved social displays—the tools of specific social motives. Like the behavioral displays of other animals, our facial movements encode information about our behavioral tendencies, not our emotional states (Andrew. 1963; Fridlund, 1990; Hinde. 1985; Smith, 1977). Some proponents of this view have interpreted audience effects as evidence that facial displays have no necessary relation to specific emotions (Bavelas et al., 1986; Fridlund, 1990; Gilbert. Fridlund & Sabini, 1987; Kraut & Johnston, 1979). We believe that developmental data are crucial to resolving this issue. Audience effects on adults' expressions might reflect learned, culture-dependent rules for facial behavior in particular social settings (Ekman & Friesen. 1982), or innate display rules, or both. Audience effects on the facial expressions of comparatively unsocialized children should be less ambiguous. We have previously reported what we believe to be the first evidence for an audience effect on smile production before the end of infancy (Jones & Raag, 1989). In that study. 18-month-old infants were videotaped playing alone while their mothers sat passively nearby. The playing infants smiled more at attentive mothers than at inattentive mothers, but otherwise gave no sign that mothers' attentiveness affected their enjoyment of the toys. Smiling at the toys, persistence in play, and rates of looking at mother were all unaffected by mothers' attentiveness. Moreover, when an attentive stranger was available, smiling was no longer reduced by mothers' inattention: the infants simply smiled at the stranger instead of at mother (Jones & Raag, 1989). These data provide evidence for an audience effect on smiling by 18 months of age. However, we can only speculate on the origins of older infants' sensitivity to the presence or absence of an audience. Eighteen-month-olds are by no means unsocialized, especially in the realm of affect (Demos, 1982; Kaye & Fogel. 1980; Malatesta el al., 1989; Malatesta & Haviland, 1985). Furthermore. 18-month-olds are at or near the start of the "word explosion"; a period of rapid vocabulary growth that may reflect the infant's grasp of symbolic communication (Nelson, 1988). It is possible that the social management of facial expressions is another form of symbolic communication that emerges at the same time (Bates, 1979; Zivin, 1982). In the study reported here, we looked for audience effects on smile production in much younger, less socially and cognitively sophisticated infants. Our subjects were 10 months of age. TenVOL. 2, NO. 1, JANUARY 1991 Copyright © 1991 American Psychological Society 45 PSYCHOLOGICAL SCIENCE I AijJicnce EtYcci on Smile month-old infanls ;i;c s(ill several monihs away Won) Iho masieiy o\' symhols. However, they had rceenlly hciiun to "lituali/e"' certain bchiiviors (voealizalions. reaching) lor apparently intentional use as eomniunieative "signs" (Bates. 1979: Harding A GolinkotT. 19791. Our question was whether these young infants also used their fueial expressions—in this ease, their smiles—as communieative signs (Malatesta & Haviland. 1985). An audience effect on smile production would suggest that they do. Thus, an audience effect would suggest that smile produetion is at least partially independent of emotion from a very early point in development.