Secondary Logo

Journal Logo

Regular Article

Emotional Expressions of Young Infants and Children: A Practitioner's Primer

Sullivan, Margaret Wolan PhD; Lewis, Michael PhD

Author Information
  • Free
  • ISEI Article


SYSTEMS for deciphering the facial expressions of infants and young children were developed in the 1980s. These systems represent important and needed tools for promoting more accurate, empirical study of early emotional development. With them, researchers are beginning to understand when infants first express particular emotion signals facially, the organization of these signals, and their relation to other aspects of motivated behavior. While many important questions about the meaning and developmental course of early expressions continue to be researched and debated, it is now recognized that most, if not all, of the facial components of the human expression repertoire can be observed shortly after birth (Camras, Holland, & Patterson, 1993; Izard & Malatesta, 1987; Lewis & Michalson, 1983). Despite this finding, facial coding systems have only just begun to move out of the laboratory into clinical settings (Gilbert et al., 1999). This is unfortunate, because information about facial expressions potentially has practical import for those working with infants, older, nonverbal children, and their caregivers.

For practitioners, facial expressions are informative in 2 ways. First and foremost, they are social signals to others. Crying, vocalizing, and bodily movements combine with facial expression to provide cues to an infant's status. Savvy caregivers no doubt make use of all of these cues in interpreting infant behavior. Although psychologists may be reluctant to assign a specific emotion value to infant facial expressions, parents have no problem doing so. In fact, they routinely use facial expressions to attribute personality and intellectual characteristics to a young baby (Haviland, 1983). They also use facial expressions to gauge their own responses, thus helping to regulate their infants' arousal and teaching them display rules (Malatesta & Haviland, 1982). Parents' reading and interpretation of the infant's expressive cues in daily interaction are key to the child's social development, emotion regulation, and early language learning (Mundy & Willoughby, 1996; Walden & Knieps, 1996). Consequently, whether parents observe and are having difficulty interpreting the expressive signals of their infant should be an important screening question for the practitioner.

Although some adults seem to have “natural” skills at reading emotional expressions, others will need some training to recognize the facial signals of young infants whose expressions are often fleeting, subtle, and perhaps less well organized than those of older children. This problem can be compounded when there is neurological impairment or developmental delay. Expressive behaviors of infants and children with various forms of disability have been described as muted, hard-to-read, or excessively labile and incongruous, depending on the population under study (Mundy, Yirmiya, & Sigman, 1990; Kasari & Sigman, 1996; Sigman, Kasari, Kwon, & Yirmiya, 1992). When parents are unable to recognize and interpret emotional signals from their infants, they will be uncertain about their child's needs and less able to share positive affect. Consequently, parents may become less expressive themselves (Dawson, Hill, Spencer, Galpert, & Waton, 1990), further degrading the quality of interaction. Learning to recognize facial expressions and how they evolve is a worthwhile effort because these and other nonverbal cues of emotion offer another channel of communication when children's behavior or vocal utterances are absent, or unclear. This channel signals the infant's arousal level, something about the quality of the infant's positive or negative response to concurrent stimulation, and may help explain other aspects of motor, postural, and behavioral responses. In short, the ability to interpret infants' expressive signals is extremely important to parents personally and to promoting mutually satisfying interactions.

Another way expressions can be helpful to practitioners is that they offer clues to the neurological and cognitive status of the child. Facial expressions are controlled through the facial cranial nerves but are also intimately linked to cognitive development (Lewis & Michalson, 1983). Consequently, young children's facial expressions have some clinical significance. Although their initial appearance is organized at the brain's subcortical level, changes in their form or developmental pattern over time reveals that the child's higher cognitive and motivational systems are becoming integrated as brain maturation proceeds. Appropriate developmental changes in facial signals imply that certain cognitive functions are preserved in neurologically damaged children, for example. This fact has long been recognized for smiling (McCall, 1972), but is likely to be true for other expressions as well.

Data on individual differences in facial expressions, their developmental trajectory in normally developing populations, and specific information about facial expressions in atypical populations is growing. This article surveys the expressions of which infants and young children are capable as revealed by facial coding systems such as MAX (Maximally Discriminative Facial Coding System), its companion, whole-face scoring system, AFFEX (Affective Expressions Scoring System; Izard, 1982), and Baby FACS (Facial Action Coding System; Oster, 1978) and will cover typical expressions. The article will not treat related and important topics such as arousal, physiological reactivity, or emotion regulation. Rather, the focus is a pragmatic one of how to recognize expressions—whether they are muted or intense, prototypic or idiosyncratic. Likewise, we set aside the thorny theoretical question of whether the emotional experiences of very young children are similar to or different from those of older children and adults. Since the discussion of expressions will make use of the movements described in MAX, AFFEX, and FACS, we briefly consider the nature of these tools.

Although some of the assumptions underlying MAX and AFFEX, as opposed to the Baby FACS coding systems, remain controversial, research using these systems shows that most of the facial movements comprising adult emotion expressions are present and map onto positive and negative reactions to stimulation during the first months of life. MAX, AFFEX, and Baby FACS also share many features. For example, all observe movements of the facial musculature in the forehead, eye/cheek, and mouth regions of the face. They also agree, for the most part, on those facial movements that constitute emotion signals. The major differences between the systems are their theoretical orientation and comprehensiveness in scoring facial movements.

MAX and AFFEX focus on a theoretically limited set of “prototypic” expressions of adults. In contrast, Baby FACS does not, and is comprehensive in scoring all possible movement combinations. Because MAX (Izard, 1983/1995) concentrates on those facial movements that are maximally discriminative of 9 specific human emotion expressions, it lends itself more readily to applied situations. Besides the 9 facially expressed emotions (ie, Interest, Surprise, Enjoyment, Anger, Sadness, Fear, Disgust, Contempt, and Shame), several control or regulatory movements are also included in MAX. Blended expressions, describing combinations of specific facial movements in the face, also are deciphered (for example, Anger/Sad is a common expression blend in infants). In MAX, single component expressions are thought to reflect a lower intensity of expression than are signals occurring in all 3 facial regions (Abe & Izard, 1999). MAX, like Baby FACS, is designed for use with videotape or still photographs, but AFFEX can be used as a live observational system. If you are interested in what facial expression or expressions can be “read” as emotional signals, MAX or AFFEX is the best application. We use it here to present the facial expressions of infancy, describing specific expressions, their development, and what is currently known about them in specific disabling conditions. The expressions are described as related sets or families because of the well-established finding that infants' expressions to any given context vary across individuals. Even within individuals, multiple expressions seen in rapid succession are common. Expressions are also seldom pure and may blend 2 emotion signals. Since blended expression may persist in some children (Yirmiya, Kasari, Sigman & Mundy, 1989), the developmental course of and individual differences in blends are of some interest.


Some do not consider interest is an emotional expression, but since it is a sign of positive approach and receptivity to people and objects, it is coded in MAX/AFFEX as a family of expressions distinct from a neutral, awake, but nonexpressive face. Figure 1 illustrates 3 varieties of interest occurring in young infants. Notice that in all of the photographs, the infants appear alert and attentive. However, there are subtle differences in their expressions, signaling differing qualities of that attention.

Fig 1
Fig 1:
Three forms of interest: (a) open interest, (b) knit brow interest, and (c) interest with lip roll.

Figure 1a has been characterized as the “open” or relaxed interest expression (Sullivan & Lewis, 1989; Sullivan, Lewis, & Alessandri, 1992). The brows are raised slightly, eyes wide open. The mouth is relaxed and open, taking a bow-shape as shown here. The mouth may be closed in some variants, but the key is that there is no sign of tension. This expression occurs in environments that offer the infant low intensity, nonthreatening stimulation. Also described as “curious” and “wide-eyed wonder” in the nonscientific literature, open interest is by far the most common expression of young infants. Adults, on the other hand, typically maintain a neutral/awake expression as our modal expression. This is not the case for infants, whose faces are rarely still and whose expressions change rapidly (Malatesta & Haviland, 1982). In fact, a “neutral” or sober face is rare in awake, attentive infants much before about 9 months of age (Lewis & Sullivan, 1988). The common occurrence of open interest reflects the “positivity offset” of the central nervous system (Cacioppo & Gardner, 1999). That is, humans show a mild bias toward positive emotion and a motivation to approach novel objects, stimuli, or contexts. Very young infants showing the relaxed, open interest expression are ready to explore—if only visually—the environment around them.

Figure 1b shows a form of interest suggesting greater intensity. It is “excited” or “knit-brow” interest (Camras, 1992; Sullivan & Lewis, 1989). The major difference from Fig 1a is the upper face region. The brows stand out more prominently because the forehead muscles have been contracted. They are pulled together, sometimes slightly as shown here, sometimes more strongly, making central bulges. The eyes appear slightly narrowed, resulting in crinkling or furrowing of the outer eye muscles known as crow's feet. Infants showing this expression appear to be intensely interested. They may vocalize while showing this expression or momentarily become still. They have a concentrated or even a quizzical look, as if they are actively studying what they are looking at. Knit brow interest is observed during young infants' social interactions, marking periods of gazing at mother or father (Malatesta & Haviland, 1982; Oster, 1978). It persists as much as 10 seconds, thereafter either resolving into a smile or, in some cases, fussing, if the interaction has been overwhelming (Oster, 1978). This expression has a specific developmental trajectory (Lewis, Sullivan, & Alessandri, 1990; Malatesta & Haviland, 1982). Seen frequently in young infants, it decreases between 2 and 8 months, appearing again regularly by 10–12 months (Malatesta & Haviland, 1982; Malatesta, Culver, Tesman & Shipard, 1989; Sullivan and Lewis, 1988). Knit-brow interest at these ages typically occurs in situations that challenge infants or requires problem-solving. It has also been referred to as a wary face in studies of response to novel objects. In such contexts, this expression is coupled with the inhibition of motor behavior (Bronson, 1972; Lewis & Michalson, 1983). Wariness suggests a state of heightened vigilance or uncertainty. Collectively, the findings suggest that knit brow interest is a form of highly focused, effortful attention associated with active information processing. The age change in its occurrence may mark a developmental shift in the type and quality of stimulation that infants must “work” to assimilate.

Figure 1c shows an interest expression that includes one of MAX's regulatory movements, a thinning and rolling inward of one or both lips. This expression was seen widely some years ago as the whimsical expression of Cabbage Patch@ baby dolls. This lip movement seems to be regulatory because it frequently suppresses smiling; however, its exact signal value in infants remains unclear. In older infants and children however, it tends to be combined with or precede gaze aversion. After about 12 months, coordination of this mouth expression with head and eye movements signals the child's awareness of the other's unwanted attention in social situations. Head lowering and gaze aversion function to withdraw from or reduce undesired social interaction, and when combined with this lip movement may signal either embarrassment or shyness toward a social partner, depending on the context and presence of other bodily cues (Lewis & Brooks-Gunn, 1979; Lewis, Sullivan, Stanger, & Weiss, 1989). After the onset of self-cognition at about 18 months of age, this lip expression may occur in evaluative settings. When coupled with body collapse, mouth components suggesting sadness, and/or withdrawal from a task, it signals shame or evaluative embarrassment (Lewis, Alessandri, & Sullivan, 1992; Lewis, 2000).

Infants with neurological impairment and various disabling conditions are frequently reported to have difficulties attending. These difficulties are often noted during infancy or are detected as learning disabilities as children reach school age. Differences in the form or patterning of interest expressions may appear as well and may offer early markers. Attention, unfortunately, has been studied almost exclusively as motor behavior, ie, visual, auditory, or even cardiac orienting to stimuli, not as facial expressions. Exceptions are studies of emotion during joint attention primarily in children with autism and mental retardation. In this work, children with these disabilities are compared to each other and with an MA-matched group to observe differences in MAX-coded positive, negative, interest (all forms) versus neutral expressions (e.g. Kasari et al, 1992; Yirmiya et al., 1992). This work finds that autistic children display primarily neutral and interest expressions for somewhat longer periods than do other groups. Increased interest in autistic children appears at the expense of less enjoyment expressions (Yirmiya et al., 1989). No published study has examined developmental trajectories or examined differences in interest expressions in any atypical population. Knowledge about the range, distribution, and pattern of these expressions may be of some clinical use in identifying those in need of further assessment.


The Surprise expression is rarely seen in young infants, especially in the full form shown in Fig 2. In surprise expressions, the brows are raised and prominently arched. The eyes are widened so that the white of the sclera is more evident than in relaxed interest. The mouth gapes with jaw slackened, assuming an “o” shape, and may be accompanied by abrupt, momentary stilling of other ongoing behavior. The expression in Fig 2 occurred when a 5-month-old accidentally turned on slides and taped music by tugging on a ribbon attached to her wrist (Sullivan & Lewis, in press). Surprise is an appropriate expression in this context of sudden, unexpected exposure to an audiovisual event. Yet, investigators have been stymied in their study of surprise because it does not consistently occur in situations that adults imagine would surprise babies. A jack-in-the-box does not produce surprise expressions in a majority of babies, for example. Occasionally, surprise occurs in situations where the experimenter did not anticipate observing it (Bennett, Bendersky, & Lewis, 2002; Camras, 1992). When surprise does occur, it appears briefly before resolving into some other expression—either interest, smiling, or a negative expression. Most researchers accept that infants show at least mild surprise expressions to novel events by 6 months, and some may do so sooner (Bennett et al., 2002; Charlesworth & Kreutzer, 1973). But, there are wide individual differences in whether babies display this expression even within the standard laboratory situations, suggesting that surprise may occur only in the most emotionally reactive infants. Besides the suddenness of the stimulus onset and its intensity, a key factor in surprise seems to be whether the stimulus event was expected. For instance, infants who learned that pulling a string turned on a slide with music expressed surprise at this contingent event only when first learning this response. Once learning had occurred, surprise expressions were no longer observed, suggesting that the infants now expected something when they pulled (Sullivan & Lewis, 1989). Surprise expressions in very young infants also appear to grow more intense across the first several repeated, sudden stimulus presentations. This is less characteristic of surprise in adults, who if surprised by a stimulus more than once, rapidly show an attenuated response.

Fig 2
Fig 2:
Surprise in a 5-month-old.

Like interest expressions, information about surprise expressions is lacking for various populations of infants and children likely to be seen in clinical practice. Given its relation to arousal modulation in response to unexpected events and its tendency to habituate with familiarization, information about surprise expressions in various groups is of considerable interest in assessing children's emotion regulation and emerging cognitive skills.


Because they are so readily recognized and such widely acknowledged milestones of social behavior, smiling and laughter, the 2 major facial expressions of enjoyment, have been the most intensely studied expressions in infants. Their developmental trajectory and links to cognitive processing in infants have been extremely well documented in full-term and preterm infants, as well as in infants with Down Syndrome, autism, and blindness (Calhoun & Kuczera, 1996; Carvajal & Iglesias, 1997; Cicchetti & Sroufe, 1978; Kasari, Mundy, et al, 1990; Kasari, Sigman, et al., 1992; McCall, 1972; Sroufe & Waters, 1976; Sroufe & Wunsch, 1972; Vine, 1973). While smiling appears universally, cultural and environmental differences also have been described (Camras et al., 1998; Gerwirtz, 1965; Kisilevsky, et al., 1998). Some have argued that additional enjoyment forms are detectable in older infants (Fogel et al., 2000; Scanlon-Jones et al., 1990). In fact, the developmental course of enjoyment provides a model for comparison with the development of other expressions, although not all expressions will follow enjoyment's pattern.

Smiling, like interest, is present from the opening days of life and is state-dependent during the neonatal period. Unlike the interest expressions, which presume an awake, alert infant, the first signals of enjoyment appear during the newborn's sleep. Even at this early stage, newborn smiles include the 2 most recognizable components of this expression: narrowed eyes, widened mouth, with corners raised. These same features consistently appear in the smiles of older infants (see Fig 3a–c), although later smiles are more intense and may include other components. Originally thought to be related to digestion (older literature may refer to them as “vegetative” or “gas” smiles), sleepy smiles are probably related to the discharge of pleasant stimulation of some kind by the infant's immature central nervous system (CNS) during rapid eye movement sleep. The source of this stimulation need not be internal, but can be linked to the external environment through whatever attention and perceptual processing systems are functional at birth. For example, one of the author's newborn daughter slept in a room with a musical, chiming clock on her first day at home from the hospital. Two days later, when she was again sleeping nearby, the clock chimes elicited smiling. This incident in a healthy newborn shows that the CNS is primed to recognize familiar external auditory information and to signal this recognition to others. Repetition alone may be enough to promote positive emotional responses to nonthreatening stimulation through simple associative learning (Zajonc, 2001).

Fig 3
Fig 3:
Four examples of enjoyment from 2- to 9-month-olds in various contexts. (a) Response to tactile teasing at 8 weeks, (b) response to audiovisual contingency at 4.5 months, (c) response to play in an older infant, and (d) response to audiovisual contingency in an older infant with Down syndrome.

Between 6 and 8 weeks, smiles of enjoyment can be observed during waking to both visual and auditory stimulation. Visual stimuli alone elicit smiling in young infants but must have a face-like quality. For example, a gently bobbing oval with 2 small black circles, suggesting the eyes of a human face, are sufficient to make a 2- to 3-month-old smile (Ambrose, 1963; Vine, 1973). Figure 3a shows that gentle tactile stimulation also successfully elicits enjoyment in an 8-week-old. Such demonstrations show that expressions of enjoyment in early life are closely tied to the physical quality of stimulation, which can be auditory, tactile, or visual, but is usually multisensory in quality. Very soon, however, enjoyment begins to occur to social events.

Expressions of enjoyment to social stimulation increase dramatically by 16 weeks. Social smiling peaks between 12–14 weeks of age in home-reared infants in Western culture. Infants at this age clearly seem to enjoy people and will smile readily at most adults who interact pleasantly with them. After 16 weeks, however, many infants become more discriminating about whom they will smile for. The familiarity and the behavioral style of the interactive partner become important factors. In this way, an emotional signal provides an index of the infant's growing social-cognitive development.

In 12- to 17- month-old infants, variants of social smiling have been described (Fogel et al., 2000; Scanlon-Jones et al., 1990). These appear to signal somewhat different qualities of enjoyment within social situations but more work is needed to determine if they are actually different expressions or just differing intensities of enjoyment (see Table 1). None are treated as unique emotions in MAX/AFFEX, but are simply coded as enjoyment. All of these expressions were more frequent during interaction with mother than with toys. The play face, however, has special meaning.

Table 1
Table 1:
Variants of enjoyment

The laughing or play face

Wide-open mouthed, gaping enjoyment expressions occur with characteristic vocalizations known as laughter. In intense forms of this expression, the lips may be rolled inward as the jaw gapes. Either form signals high excitement and positive arousal. Such expressions haven been called “play face” because they appear to be the human equivalent of a primate expression of the same quality (Blurton-Jones, 1972; Preuschoft & van Hoof, 1997). The play face makes its appearance by 5 months of age in normally developing infants (Sroufe & Waters, 1976). It occurs at first to vigorous, auditory, and tactile stimulation, ie, tickling as did smiling a few weeks previously. After 7 months, visual stimulation becomes more effective in eliciting this expression. By about 12 months, incongruity and novelty, especially if it involves the infant's own participation, will elicit laughter and the play face, immediately or in anticipation of game-like interaction with a social partner, such as peek-a-boo (Fogel et al., 2000).

Enjoyment of mastery

During the second half of the first year, infants continue to smile and laugh at nonsocial events but infrequently compared to social situations. Smiling occurs in nonsocial contexts especially when infants learn that interesting, familiar, and therefore, nonthreatening stimulation is under their own control. Figure 3b shows a 4-month-old expressing enjoyment of a slide and music that she has learned to expect when she tugs a ribbon attached to her wrist (Lewis et al., 1990; Sullivan & Lewis, 1989). This is the beginning of playful enjoyment of mastery of the physical environment that will continue to develop as infants learn to play and explore. Comparing this expression to one of the 9-month-old at play with something novel (Fig 3c), one can see that the crinkled eyes, widened mouth, and raised, dimpled cheeks are very similar, despite the fact that motor and cognitive skills have grown considerably in the intervening 5 months. Thus, it seems that the form of the enjoyment expression does not change, only the contexts and qualities of stimulation required to elicit it. Enjoyment of mastery of an object or task becomes especially common after 12 months and has been studied as one measure of “mastery motivation” (Busch-Rossnagel, 1997). Once the infant is capable of self-referential and self-evaluative behavior, the integration of enjoyment of mastery and postural and gestural signals will lead to more elaborated expressions of pride.

Because of the connection of enjoyment with management of arousal and emerging cognitive competencies, smiling and laughter have been the most studied expressions of infants and children with disabilities. Much is known, for example, about the emergence of enjoyment in infants with Down syndrome (DS) (DS; Cicchetti & Sroufe, 1978; Sroufe & Wunsch, 1972). As can be seen in Fig 3d, the characteristic expressive components of mastery enjoyment are present in this 8-month-old child. Research shows that the general course of both smiling and the play face is the same in infants with DS, although delayed compared to infants without this syndrome. For example, social smiling peaks at the same mental age (ie, 4–5 months) and smiles are directed appropriately to social partners (Carvajal & Iglesias, 1997). The intensity of enjoyment is less, however, and infants with DS may not sustain social enjoyment in spontaneous interaction to the same degree as do infants without the syndrome and may be less likely to initiate smiling (Carvajal & Iglesias, 2000; Kasari & Sigman, 1996). Such differences can have important effects on social interaction. Smiling to auditory stimulation also may be more limited in some children with DS, but can be increased with intervention (Calhoun & Kuczera, 1996). In fact, by preschool age, children with DS are usually reported to be highly positive in their social interactions. One study finds however that they are uniformly high in expression enjoyment across situations whereas MA-matched peers displayed more enjoyment only in those situations involving joint attention and affect sharing (Kasari, Mundy, et al., 1990; Kasari, Sigman, et al., 1992). Thus the child with DS seems to display enjoyment more indiscriminately, suggesting a need to focus on emotion regulation in some contexts. Finally, enjoyment of mastery may be affected in some children with DS because motor limitations restrict early forms of object play, but this also can be improved by intervention (Brinker & Lewis, 1982; Sullivan & Lewis, 1990).

In severe cerebral palsy (CP), enjoyment expressions, especially in the mouth region, may be less well modulated or appear asymmetrical. If motor impairment is severe, the play face may be the only form of smiling available to these children because subcortical centers modulating enjoyment are activated.

The course of smiling in blind children also has been studied to understand the degree of visual input that is necessary to produce recognizable expressions of happiness. Visual input does not appear to be necessary for blind children to produce recognizable, spontaneous expressions of enjoyment. Instead, it is their negative expressions that are more difficult to recognize compared to those of sighted children (Galati, Sini, Tinit, & Miceli, 2001). However, blind children may be less skillful at management of social expressions, contributing to the social inhibition that is frequently reported in these children (Castanho & Otta, 1999). Lower rates and poorer quality of enjoyment expressions may prompt the perception of inhibition in children with disabilities generally.

Enjoyment expressions in autistic children have been studied around the issue of shared affect between child and adult partners. Less enjoyment is expressed by these children when compared with MA-matched controls in semistructured interactions. Autistic children also fail to share enjoyment with social partners by coordinating their enjoyment expressions with eye contact (Dawson et al., 1990; Sigman et al., 1992). During toy play, autistic children are as likely to express enjoyment, a finding that underscores the nonsocial nature of their expressions (Sigman et al, 1992). Finally, autistic children are likely to display incongruous blends expressions that combine positive and negative facial elements (Yirmiya et al., 1989). It is unclear if this occurs because of problems in regulating expressions themselves, or because of ambivalent or competing emotional arousal in these children. However, incongruous expressions provide unclear, conflicting signals to caregivers who rely much on context or trial and error in responding to them.


Some argue that pain is not an emotion. Yet, painful stimulation clearly causes a strong negative emotional response and promotes other negative expressions. The developmental course of pain expression has been studied in some detail because of its theoretical interest and the very practical need for assessment and management of pain in pediatric procedures (Grunau, Oberlander, Holsti, & Whitfield, 1998; Oberlander, 2001). Pain expressions can be also observed in situations of distress that are not physically painful (Oster, Hegley, & Nagel, 1992). Thus, the pain expression and those that follow it provide clues to emotional and regulatory responses to all forms of aversive stimulation.

Acute pain in response to tissue damage during standard pediatric procedures (eg, circumcision, heel lance, or inoculation) provides a naturalistic and ethical way to observe how facially and behaviorally expressed pain responses change with the developmental and neurological status of the infant. Pain in response to such procedures is signaled by distinctive and intense facial actions including the drawing together and lowering of the brows to create a midbrow bulge, a deepened nasolabial furrow, and tight squeezing of the eye orbit muscles, resulting in a strong squint. From the newborn period through 18 months, few changes occur in the pain expression's appearance or components (Craig, 1992; Craig & Grunau, 1993; Izard, Hembree, Dougherty, & Spizziri, 1983; Johnston, Stevens, Craig, & Granan, 1993; Lilley, Craig, & Grunau,1997). Throughout this time, the facial response to acute pain reliably includes all of the upper face movements listed. Mouth movements are more variable but include lateral stretching of lips, especially in older infants and children. In young infants, one of two common mouth variations can be observed. Prominent especially in newborns is a dropped jaw with taut or "cupped" tongue within an angular, wide mouth (see Fig 4a; Lilley et al., 1997). From 1 to 5 years, these facial movements cohere to form pain expressions in pediatric patients. A greater number of components shown is related to higher pain ratings by clinicians (Gilbert et al., 1999). The laterally stretched mouth also occurs commonly at these ages (see Fig 4b). Since few longitudinal investigations of pain have followed infants' pain expressions from the newborn period, the meaning of these variations and age changes in the mouth components of pain are unknown. We do not know if they reflect individual differences in pain sensitivity or in regulatory responses to pain.

Fig 4
Fig 4:
Pain in response to DPT inoculation in 4-month-olds showing (a) cupped tongue and (b) lateral retraction of the mouth.

Surprisingly, increased crying in preterms and newborns is not a reliable marker for pain in response to a heel lance (Grunau & Craig, 1987). Young babies vary in their irritability and many will cry in response to handling prior to the actual procedure (Grunau, Johnston, & Craig, 1990; Owens & Todt, 1985). Very low birth weight premature infants between 26–31 weeks gestational age, show the upper face pain actions when their heels are lanced to obtain blood. The upper facial response is specific to the piercing of the skin, rather than other potentially stressful aspects of handling that occur as part of the medical procedure, and is accompanied by the maximum increase in heart rate. Noxious stimulation and the high negative arousal they produce appear to simultaneously activate many different negatively toned neurological systems in the young baby. Pain expressions are associated with a rise in cortisol levels in newborns, also suggesting that heel lancing is a highly stressful procedure for the young baby (Oberlander, Gilbert, Chambers, O'Donnell, & Craig, 1999; Owens & Todt, 1985; Ramsay & Lewis, 1994). The pain expression and its accompanying physiological response are related to the developmental age of the preterms, appearing more consistent and robust in older babies (Johnston, Stevens, Yang, & Horton, 1995). It is unclear if this age change reflects better neurological regulation of the pain response, or the gradual recovery from illness and trauma experienced by many of these sick babies.

Although the expression of pain appears relatively invariant over the first 2 years of life, a number of important changes occur that possibly reflect a combination of neurophysiological maturation, life experience, and a growing ability to remember prior painful experiences. In contrast to newborns, in 2-month-olds, the expressive components of pain occur at low frequency during a preinoculation or baseline period. This observation suggests that pain expressions, occurring as nonspecific generalized distress reactions to handling, decrease with age. However, because a significant and dramatic increase in all pain components is observed in response to inoculation, infants' pain response shows some specificity to skin trauma at every age studied (Lilley et al., 1997). Healthy term infants between 2 and 4 months of age have the most robust response to pain (Lewis & Thomas, 1990; Maikler, 1991). By 4 months, pain expressions are highly specific to inoculation, with very few pain signals occurring during the baseline period. Likewise, 4 month-olds have quicker recovery from pain (Lilley et al., 1997; Lewis & Thomas, 1990; Ramsay & Lewis, 1994), suggesting that CNS mechanisms inhibiting the transmission of pain become functional at this age. Following this important transition point, 6-month-olds show a shorter duration of pain response and less of a rise in cortisol in response to immunizations, suggesting better internal physiological regulation in response to pain (Lewis & Thomas, 1990; Ramsay & Lewis, 1994).

Although the appearance of the pain expression may change little with age, older infants have more complex responses to pain. Typically, they display facial pain for a smaller proportion of time prior to quieting, displaying anger and blended expressions instead (Izard et al., 1983; Izard, Hembree, & Heubner, 1987). By 18 months, pain-specific expressions comprised only 10% of the postinoculation distress (Izard et al., 1987). Thus, anger and other negative expressions become rapid after-reactions to the initial pain response. This pain after-reaction is most likely to influence the appropriate soothing strategy and might also be stable across individuals. For example, these postpain facial signals likely reflect some combination of differences in pain sensitivity and social experience among individuals. For, example, Japanese infants seem to have less pain sensitivity and qualitatively different emotional responses to inoculation than do American infants; pain expressions are less intense and are not typically followed by anger and crying, but by surprise (Lewis, Ramsay, & Kawakami, 1993). Differences between Chinese infants and other groups are even more pronounced for a variety of expressions (Camras et al., 1998).

Greater behavioral coordination and maturation of motor skills also lead to elaboration of responses to pain in older infants. Infants over 12 months of age are more likely to show goal-focused behaviors following immunizations (withdrawing, avoiding), compared to younger infants who display more unorganized distress responses (Craig & Grunau, 1993). Between 12 and 18 months, pain responses become anticipatory. Lilley et al (1997) reported that baseline pain expressions were low between 4 and 12 months, but rose again between 12 and 18 months to a level similar to that of 2-month-olds! These differences in the baseline or pre-inoculation period suggest that the older infants anticipated the painful procedures in store for them. If so, pediatricians may begin to encounter anticipatory emotional responses to pain, cued by the environment and salient memories of prior procedures by 12 months, perhaps even earlier. Collectively, these changes suggest that the pain expression is stable over the first 2 years of life and there is a close link between facially signaled pain and neurophysiological development.

Significant neurological impairment appears to dampen the pain response, but the basis for dampened or atypical pain responses is unclear (Oberlander et al., 1999). Significant neurological, motor, and/or cognitive impairment can produce a decreased pain response for any number of reasons, leading to underrecognition and poor management of pain in children with disabilities or neurological impairment. The development of expression-based pain scales for pediatric practice is a relatively recent application directed toward addressing this problem and discovering methods to assess pain based on facial responses (Oberlander, 2001). Hopefully, better quantification of pain will lead to better recognition and effective management.


The disgust expression is another that has been studied in some detail in newborns. The newborn's response to bitter and sour substances is distinct from responses to water and sweet solutions (Granchow, Steiner, & Daher, 1983; Rosenstein & Oster, 1988; Steiner, 1979). Saltiness does not trigger disgust expressions, with less than half of infants showing any negative response to it (Rosenstein & Oster, 1988). However, quinine and other bitter tastes are potent and rapid elicitors of disgust responses regardless of infant state. The intensity of the disgust expressions as well as the pattern of components expressed appear to index increasing concentrations of the solutions presented (Granchow et al., 1983; Rosenstein & Oster, 1988; Steiner, 1979). The response to a bitter taste typically involves turning away from the source of the stimulus and gaping of the mouth and lower lip, as shown in Fig 5a. Sometimes, the gag reflex is observed (Steiner, 1979). Nose wrinkling and upper lip raising are components of the full disgust reaction, but occur alone as observed in about half of the newborns as a less intense form of the reaction (Rosenstein & Oster, 1988). A flattened tongue and drooling may also be visible, depending on the state of the child and the concentration of the fluid (Steiner, 1979). These responses signal strong distaste on the part of the infant. They are a defensive reflex helping the infant to rid itself of the unappealing taste. Similar responses are observed to odors that most adults would find objectionable, such as fishy and “rotten egg” smells (Steiner, 1979).

Fig 5
Fig 5:
Disgust (a) and sour (b, c) responses of 4-month-old infants elicited by a lemon juice swab.

Responses to a sour taste, such as citric acid solution, are more variable, seem to be milder, and evolve over several seconds in contrast to disgust. Lip pursing is the commonly observed initial response, accompanied by nose wrinkling, narrowed eyes, and blinking (Rosenstein & Oster, 1988; Steiner, 1979). Figure 5b shows this response in a 4-month-old. A closed mouth, either with corners down or retracted lips, is a feature associated with milder aversive reactions. This “sour grimace,” signaling dislike, occurs as the response develops (Granchow et al., 1983; Steiner, 1979).

The variability of sour expressions in response to sour tastes is even more pronounced in 4-month-olds than in newborns. The most common response to a natural sour taste (lemon juice) at this age is interest, following the initial puckering and rapid lip movements associated with tasting or mouthing of the flavor. In some infants, however, negative expressions indicating dislike (sad frown or lip retraction movements) appear, as observed in newborns (see Fig 5c). However, a few infants actually smile, the next most common response to this taste (Bennett, et al, 2000). This finding shows that early individual differences in taste preferences and rapid resolution of initial mild disgust reactions into either enjoyment or dislike emerge early in life.

The developmental course of the disgust family of expressions has not been traced further, but it appears that the disgust response is so well-developed in infants and specific to both the quality of olfactory and gustatory stimulation, that its form changes little. Steiner (1979) undertook extensive study of disgust and enjoyment taste reactions in various populations of infants and children with various forms of disability, including newborns with hydrocephaly, anencephaly, the congenitally blind, children with craniofacial malformations, and mentally retarded adults. Disgust and sour grimace expressions were recognizable in all populations studied, despite the considerable variation in the cognitive and motor control (Steiner, 1979). He argued that the preservation of disgust responses across this wide range of cognitive and motor functioning means that disgust expressions are controlled by the brainstem and so undergoes little morphological change. However, the emergence of the positive or negative reactions to sour also points to 4 months as an important developmental milestone in organization of expressions.

Perhaps the most interesting aspect of disgust expressions, and the one about which little is known, is how the subtle variations in responsiveness observed even in newborns are related to individual differences in nervous system functioning or other factors, such as temperament. A second unanswered question is when these expressions begin to occur in response to non-taste stimuli, and especially social stimulation. Like smiling, disgust expressions are elicited initially to physical rather than social-psychological elicitors. Non-taste stimuli that might be sufficient to produce disgust in infants have not yet been reported, but can be imagined. Disgust signals stimulus rejection. Therefore, too rapid or sudden occurrence of a stimulus might elicit disgust expressions if the stimulus overwhelms infants' ability to process it. We have observed this on some occasions in the infant learning lab: sometimes an infant pulls sufficiently rapidly that the slide and music appears within a second of its previous appearance and before the infant's reaction to the first appearance has subsided. When this occurs, the infant may display nose wrinkling, or asymmetrical mouth and lip movements in response, suggesting a kind of recoil reaction to the overwhelming, too rapid reoccurrence of the stimulus. While clearly not yet an instance of “psychological disgust,” such observations suggest that prior to 6 months of age, components of disgust may occur in response to visual and auditory stimulation, setting the stage for their later function in social situations. But, the forms of disgust observed in this case do not involve the intense, gaping reactions observed in response to bitter tastes.

Toilet training and introduction of solid foods when the child is able to eat independently are thought to be important contexts for parent-child socialization of disgust (Rozen, Hadt, & McCauley, 2000). It is also possible that disgust, signaling stimulus rejection, is more common in neurologically damaged infants and children in whom inhibitory control is slow to develop or absent. This response would be an important clue to caregivers to reduce the intensity or pacing of stimulation.


The prototypic anger or cry face appears in Fig 6a. It is characterized by brows drawn together and lowered, sometimes strongly as seen in this view. Deep nasolabial folds frame a wide-opened square mouth. Anger expressions of this intensity are almost always accompanied by a rolling cry in infants, but milder versions are observed without any vocalization. This expression is the infant's most common negative expression.

Fig 6
Fig 6:
Anger (a) and upper face anger blended with sad (b) expressions in 5-month-olds.

A major quandary for theorists is that the anger/cry face rarely occurs in young infants without the co-occurrence of other negative expressions or blends. Blended anger and sadness, as seen in Fig 5b, is commonly seen. In this expression, knit, lowered brows and narrowed eyes of the upper face region are accompanied by a sad "pout" in the lower face. This particular blend (upper face anger with lower face sadness) is often observed during social interaction with the mother and it tends to decrease with age in this setting (Izard et al., 1995). This might not be so in all situations that recruit anger expressions, but more information is needed. The co-occurrence of multiple negative and blended expressions with the anger/cry face has led some to argue that these negative expressions collectively reflect general distress or unhappiness instead of a specific facial signal of anger or other negative emotions (Camras, 1992). However, the co-occurrence of multiple negative expressions in the early months can also be explained by the well-known quality of the CNS to respond more intensely to negative than to positive stimuli (Peters & Czapinski, 1990). Although the CNS idles in a mildly positive mode (positivity offset), when a threatening stimulus is encountered, a vigorous negative response is observed. This phenomenon, known as the negativity bias, may be especially evident in young infants, in whom inhibitory control is still limited and arousal is poorly regulated. The negativity bias could result in an initially unmodulated negative response that simultaneously activates several competing systems when an aversive event is initially perceived. It is not that there is one “undifferentiated” distress state, but that all negative emotion systems are primed. The strongest response tendency is an energizing response to action in order to resist or overcome the aversive stimulus. It is signaled by increased motor activity and crying as well as by the prototypical anger or cry face expressions.

Although anger expressions may not appear exclusively when infants encounter certain types of negative events, by 3–4 months of age, a number of situations seem to result in anger expressions predominantly. Restraining an infant's arms, for example, produces a variety of facial expressions, not all of them negative. Anger, however, is the predominant negative expression (Bennett et al., 2002; Braungart-Reiker & Stifter, 1996; Stenberg, Campos, & Emde, 1983; but see Camras et al., 1998, regarding ethnic differences). An even more powerful example is the response of infants to loss of a contingent event (Lewis et al, 1990; Sullivan, Lewis, & Alessandri, 1992). Infants who learn to control a pleasant event by pulling a ribbon show anger expressions and increase pulling when that action abruptly fails to produce the event. Anger expressions increase, but other negative expressions do not. The increase is therefore specific to infants' having learned a relation between action and outcome (Lewis et al., 1990; Sullivan & Lewis, in press). Situations that produce anger expressions before 6 months of age are those in which access to desired objects or goals is blocked or thwarted in some way (Bennett et al., 2002; Lewis et al., 1990; Stenberg et al., 1983; Sullivan et al, 1992) or, those causing pain after about 4 months (Izard et al., 1983). There is stability in individuals' expressions of anger in these contexts as well (Izard et al., 1995; Sullivan et al, 1992).

Crying and fussing are widely recognized vocal signals of negative emotion in infants, but no any one quality of crying appears to be specific to the “cry face” or to any other negative expression. Negative facial expressions can precede negative vocal behavior, suggesting that greater or increasing arousal is needed for vocal crying. Cries can be differentiated by frequency and tonal quality but it is difficult for observers to discriminate these features reliably and experience is an important factor when differentiating vocal crying in the mild to moderate range (Papousek, 1989). Facial expressions probably provide more graded information about the onset, quality, and intensity of negative reactivity, making such discriminations possible.

The form, development, and regulation of anger expressions and their relation to vocal crying in atypical development has not been addressed in any significant detail. Like smiling, negative vocalizations undergo an important developmental transition in the first 4 months (Hopkins, 2001). Presumably this change parallels changes in the patterning of anger and other negative expressions, although this has not been studied. Much available work has focused on the clinical utility of vocal cry behavior and unexplained infant irritability (Barr, Hopkins, & Green, 2000). Because adults rely on vocal and facial information before correctly identifying that infants are indeed upset (Greene & Gustafson, 2001), the lack of information about the coordination of expression and voice in the development of typical and atypical children is frustrating. Disabling conditions significantly affect adult perceptions of young children. For example, infants and children with DS are reportedly less irritable than the norm and those with CP and auditory and visual handicaps even more so (Field, 1996). Autistic children are also sometimes found to be more negative (Yirmiya et al., 1989). Greater attention to the quality of expressive behavior of children with disabilities might provide more clinically useful information and suggest possible interventions. Given their physical limitations, it is reasonable to expect greater frustration on the part of these children. Whether it is expressed as anger, other negative expressions, or undifferentiated distress is not known


Figure 7a shows the sad frown or “pout face” expression. The brows are raised and angular in appearance over narrowed eyes. The nasolabial folds appear prominent, as they do in all negative expressions. The mouth corners are down-turned in a “horseshoe” shape (Oster, 1978). The chin is raised, sometimes prominently as in Fig 7b. The latter movement, especially when combined with forward projection of the lower lip, as in Fig 7b, gives it the pouting quality. In both examples of this expression, the mouth is closed, although open-mouthed sad pouts are also possible. Sad pouts are equally likely to appear as an upper face signal or as a full expression in young infants (Izard et al., 1995; Sullivan & Lewis, in press). When blended with another expression, sad brows are especially likely to combine with anger or, less often, with interest in the lower face. In either case, such blends are interpreted as a signal of dislike or unhappiness.

Fig 7
Fig 7:
Sad expressions without (a) and with (b) lower lip protrusion in infants under 6 months of age.

Sad pouts have never been observed to be specific to any stimulus or context. For this reason, little is known about them. They are not a dominant negative expression in any context that has been studied, except for pouts associated with the after-sour response (Bennett et al., 2002). Sad pouts occur at relatively low but stable levels during both social interaction and learning/frustration episodes (Izard et al., 1995; Sullivan et al., 1992). Some have suggested that this expression occurs as an infant or young child either arouses to an anger expression or as a regulatory movement inhibiting anger (Camras, 1992; Oster, 1978). It is possible that sad expressions are never completely independent of anger expressions even in adults (Barr-Zisowitz, 2000). Caregiver intervention in response to sad expressions may forestall or more effectively help dampen more intense negative response, but this has not been demonstrated.


Fear expressions are notoriously difficult to observe in infants. In adults and children, this expression involves raised and straightened brows, widened eyes with tense lower eyelids, and horizontally retracted lips. Less intense versions feature brow movements alone, possibly blended with interest or other facial movements in the lower face, such as anger or sadness. Situations that could conceivably frighten infants (eg, a visual cliff, the approach of a stranger, highly novel masks, and startling mechanical toys) do not elicit fear expressions (Bennett et al., 2002; Camras, 1992; Izard et al., 1995; Lewis & Michalson, 1983). Therefore, some suggest that infants may not be able to display fear until sometime after 7–12 months because greater experience and cognitive abilities are needed for appraisal of strange and dangerous environmental events (Izard & Malatesta, 1987). Both independent locomotion and ability to inhibit behavior may also be prerequisites of fear. For example, wariness of strangers, a mild form of fear, is not observed in a majority of children until after 7 months of age (Bronson, 1972). By 11 months, the brow components of fear do occur to a variety of aversive stimuli, but are not specific to fear-inducing toys, such as a growling mechanical gorilla (Camras et al., 1998). Fear systems in the brain do have a high degree of plasticity (LeDoux & Phelps, 2000), supporting the view that learning and experience are very important to this emotion. The question remains how little experience is needed for fear expressions to occur. Figure 8a shows the response of a 6-month-old infant to the return of a nurse for the second of 2 inoculations administered at this well-child visit. This observation suggests that a single salient experience may be sufficient. By 2 years of age, fear can be rapidly learned, resulting in phobic reactions that may spread readily to previously nonfeared objects through learning (Watson & Rayner, 1920).

Fig 8
Fig 8:
A fear expression in response to approach of a nurse following inoculation (a) and a fear blend in response to the sudden appearance of a stimulus (b).

Even in young infants, fear blends are observed on occasion in situations that are aversive. For example, fear blends occurred in 8% of infants in response to arm restraint (Camras, Oster, Campos, Miyake, & Bradshaw, 1997). This raises the possibility that most stimulus situations studied thus far are not adequate to elicit fear expressions and that blends are observed because milder stimulation results in less intense reactions. As with disgust, very specific and intense stimulation may be necessary to elicit fear expressions. Looming objects, very loud sounds, and the loss of support are good candidates. However, some of these are impossible to study experimentally because of ethical considerations. Alternatively, aversive situations may produce fear only in the most temperamentally fearful children. In our studies of infant conditioning, fear blends appear as negatively toned surprise expressions (see Fig 7b). These surprise-fear blends appear early in the conditioning session, before infants realize that their own movements cause the sudden onset of slides and music. While most infants are surprised and interested in these first appearances of slides and music, others show surprise reactions that grow more extreme across several presentations, ultimately developing a negative quality including fear components (brow straightening, tensed lower eyelid, and/or lateral movement of the mouth corners). In most cases, fear-surprise responses subside quickly as infants learn the relation between their own movements and the slide's appearance (Lewis, Sullivan, & Michalson, 1984; Sullivan & Lewis, 1989). But, some infants begin to fuss and are unable to complete the procedure. These infants are reported by mothers to have more fearful temperaments (Sullivan, Ramsay, & Lewis, 1992). It seems reasonable to infer that infants with fearful temperaments may be expressing fear.

Because so few negative situations elicit fear and the percentage of children for whom fear expressions have been observed is so low (Bennett et al., 2002), the developmental course of fear expressions has not been studied extensively. The dependence of fear on cognitive skills and appraisal of danger in the environment may mean that most infants must learn to fear. Hence, fear expressions will vary greatly across individuals. Some infants may express excessive amounts of fear, either because neurological damage has interfered with the normal buffering of the fear expression, or because prolonged experience with invasive medical procedures have resulted in conditioning of fearfulness.


Despite differing approaches to measurement, considerable evidence shows that context-appropriate emotion expressions occur in infants and young children from birth or shortly thereafter. No matter what system is used, interest, enjoyment, anger, pain, and disgust have been shown to signal behavioral dispositions congruent with these emotions in young infants. This is also true for fear and sad expressions in older infants. Pain, enjoyment, interest, and possibly anger and disgust show little change over the first 2 years. Instead, the contexts that elicit these emotions shift at several major transitions in cognitive development and cerebral maturation, ie, 4 months, 7–9 months, and 18–24 months. At these developmental ages, expressions can be particularly informative about the developmental status of children. At all ages, individual differences in expressive behavior can provide clues to children's preferences and style of response. Although a number of interesting and important questions remain about expressive development, it seems clear that facial expressions provide important information to caregivers and practitioners alike. Expressions, especially when combined with vocal and postural behaviors, provide important clues to the motivational state of infants who cannot otherwise report what they feel. Ability to read the facial behavior of children with disabilities gives parents an important tool for promoting positive emotion, better management of their children's negative behavior, and preventing frustration or helpless withdrawal of the parents. It is also clear that social environments have substantial influence on the emotional behaviors of children with disabilities as well (Field, 1996), suggesting that well-designed interventions may help parents who have difficulty in understanding their children's expression become better attuned to subtle, incongruous, or ambiguous signals. Awareness of these expressions and their developmental trajectories by the practitioner can assist in assigning meaning to infant behavior, skills that parents are especially concerned about in day to day commerce with their children.


Abe, J. A., & Izard, C. (1999). A longitudinal study of emotion expression and personality relations in early development. Journal of Personality and Social Psychology, 77, 566–577.
Ambrose, J. A. (1963). The development of the smiling response in infancy. In B. Foss (Ed.), Determinants of infant behavior (Vol. 1, pp. 179–196). New York: Wiley.
Barr, R. G., Hopkins, B., & Green, J. A. (Eds.). (2000). Crying as a sign, a symptom, & a signal. London: MacKeith Press.
Barr-Zisowitz, C. (2000). “Sadness”—Is there such a thing? In M. Lewis & J. M. Haviland-Jones (Eds.), The handbook of emotions (2nd ed., pp. 607–622). New York: Guilford.
Bennett, D., Bendersky, M., & Lewis, M. (2002). Facial expressivity at 4 months: A context by expression analysis. Infancy, 3, 97–114.
Blurton-Jones, J. (1972). Ethological studies of child behavior. London: Cambridge University Press.
Braungart-Reiker, J. M., & Stifter, C. E. (1996). Infants' responses to frustrating situations: Change and continuity in reactivity and regulation. Child Development, 67, 1767–1779.
Brinker, R. P., & Lewis, M. (1982). Discovering the competent handicapped infant: A process approach to assessment and intervention. Topics in Early Childhood Special Education, 2(2), 1–16.
Bronson, G. W. (1972). Infants' reactions to unfamiliar persons and novel objects. Monographs of the Society of Research in Child Development, 37(Serial No. 148).
Busch-Rossnagel, N. (1997). Mastery motivation in toddlers. Infants and Young Children, 9(4), 1–11.
Cacioppo, J. T., & Gardner, W. L. (1999). Emotion. Annual Review of Psychology, 50, 191–214.
Calhoun, M., & Kuczera, M. (1996). Increasing social smiles of young children with disabilities. Perceptual and Motor Skills, 82, 1265–1266.
Camras, L. (1992). Expressive development and basic emotions. Cognition and Emotion, 6, 269–283.
Camras, L., Holland, E. A., & Patterson, M. J. (1993). Facial expressions. In M. Lewis & J. M. Haviland (Eds.), Handbook of emotions (pp. 199–208). New York: Guildford.
Camras, L., Oster, H., Campos, J., Campos, R., Ujiie, T., Miyake, K., et al. (1998). Production of emotional facial expression in European American, Japanese, and Chinese infants. Developmental Psychology, 34, 616–628.
Camras, L., Oster, H., Campos, J, Miyake, K., & Bradshaw, D. (1997). Japanese and American infants' responses to arm restraint. In P. Ekman & L. Erika (Eds.), What the face reveals: Basic and applied studies of spontaneous expression using the Facial Action Coding System. Series in Affective Science (pp. 289–301). New York: Oxford University Press.
Carvajal, F., & Iglesias, J. (1997). Mother and infant smiling exchanges during face-to-face interaction in infants with and without Down Syndrome. Developmental Psychobiology, 31, 277–286.
Carvajal, F., & Iglesias, J. (2000). Looking behavior and smiling in Down Syndrome infants. Journal of Nonverbal Behavior, 24, 225–236.
Castanho, A. P., & Otta, E. (1999). Decoding spontaneous and posed smiles of children who are visually impaired and sighted. Journal of Visual Impairment and Blindness, 93, 659–665.
Charlesworth, W. R., & Kreutzer, M. (1973). Facial expressions of infants and children. In P. Ekman (Ed.), Darwin and facial expression: A century of research in review. New York: Academic Press.
Cicchetti, D., & Sroufe, L. (1978). The relationship between affective and cognitive development in Down Syndrome infants. Child Development, 47, 920–929.
Craig, K. D. (1992). The facial display of pain: Better than a thousand words. American Pain Journal, 1, 153–162.
Craig, K. D., & Grunau, R. V. (1993). Neonatal pain perception and pain measurement. In K. Anand & P. McGrath (Eds.), Pain in neonates (pp. 67–105). Amsterdam: Elsevier.
Dawson, G., Hill, D., Spencer, A. Galpert, L., & Waton, L. (1990). Affective exchanges between young autistic children and their mothers. Journal of Abnormal Child Psychology, 10, 335–345.
Field, T. (1996). Expressivity in physically and emotionally handicapped children. In M. Lewis & M. W. Sullivan (Eds.), Emotional development in atypical children (pp. 1–28). Mahwah, NJ: Erlbaum.
Fogel, A., Nelson-Goens, C., & Hsu, H. (2000). Do different infant smiles reflect different positive emotions? Social Development, 9, 497–520.
Galati, D., Sini, B., Tinit, C., & Miceli, R. (2001). Facial expressions of emotion in congenitally blind children from 8–11 years of age: Observing and encoding the expression of emotions (Italian). Research in Psychology, 23, 103–126.
Ganchrow, J. R., Steiner, J. E., & Daher, M. (1983). Neonatal facial expression in response to different qualities and intensities of gustatory stimuli. Infant Behavior and Development, 6, 473–484.
Gerwirtz, J. L. (1965). The course of infant smiling in four child-rearing environments in Israel. In B. Foss (Ed.), Determinants of infant behavior (Vol. 3, pp. 205–249). London: Methuen.
Gilbert, C., Lilley, C., Craig, K., McGrath, P., Court, C., Bennett, S., et al. (1999). Postoperative pain expression in preschool children: Validation of the child facial coding system. The Clinical Journal of Pain, 15, 192–200.
Greene, J. A., & Gustafson, G. (2001). Crying: Multiple determinants of perceived meaning. In R. Barr, I. St. James-Roberts, & M. Keefe (Eds.), New evidence on unexplained early infant crying: Its origins, nature and management (pp. 105–120). Skillman, NJ: Johnson & Johnson Pediatric Institute.
Grunau, R., & Craig, K. (1987). Pain expressions in neonates: Facial action and cry. Pain, 28, 395–410.
Grunau, R., Johnston, C. C., & Craig, K. (1990). Neonatal facial and cry responses to invasive and noninvasive procedures. Pain, 76, 277–286.
Grunau, R. E., Oberlander, T., Holsti, L., & Whitfield, M. F. (1998). Bedside application of the Neonatal Facial Coding System in pain assessment of premature neonates. Pain, 76, 277–286.
Haviland, J. M. (1983). Looking smart. In M. Lewis (Ed.), Origins of Intelligence (2nd ed., pp. 423–450). New York: Plenum.
Hopkins, B. (2001). Development of crying: The origin and change problem. In R. Barr, I. St. James-Roberts, & M. Keefe (Eds.), New evidence on unexplained early infant crying: Its origins, nature and management (pp. 71–86). Skillman, NJ: Johnson & Johnson Pediatric Institute.
Izard. C. (1982). A system for identifying affect expressions by holistic judgements (AFFEX). Newark, DE: Instructional Resources Center.
Izard, C. (1983/1995). The maximally discriminative facial movement coding system (MAX-revised edition). Newark, DE: Instructional Resources Center.
Izard, C., Fantauzzo, C. A., Castle, O., Haynes, M., Rayias, M. F., & Putnam, P. (1995). The ontogeny and significance of infants' facial expressions in the first 9 months of life. Developmental Psychology, 31, 997–1013.
Izard, C., Hembree, E. A., Dougherty, L. M., & Spizziri, C. L. (1983). Changes in facial expressions of 2- to 19-month-old infants following acute pain. Developmental Psychology, 19, 418–426.
Izard, C., Hembree, E. A., & Heubner, R. B. (1987). Infants' emotion expressions to acute pain: Developmental change and stability of individual differences. Developmental Psychology, 23, 105–113.
Izard, C., & Malatesta, C. (1987). Perspectives on emotional development. In J. D. Osofsky (Ed.), Handbook of infant development (2nd ed., pp. 494–554). New York: Wiley.
Johnston, C. C., Stevens, B., Craig, K., & Granau, R. K. (1993). Developmental changes in pain expressions in premature, full-term, 2- and 4- month-old infants. Pain, 52, 201–208.
Johnston, C. C., Stevens, B., Yang, F., & Horton, L. (1995). Differential response to pain by very premature neonates. Pain, 61, 471–479.
Kasari, C., Mundy, P., Yirmiya, N., & Sigman, M. (1990). Affect and attention in children with Down Syndrome. American Journal of Mental Retardation, 95, 55–67.
Kasari, C., & Sigman, M. (1996). Expression and understanding of emotion in atypical development: Autism and Down Syndrome. In M. Lewis & M. W. Sullivan (Eds.), Emotional development in atypical children (pp. 109–130). Mahwah, NJ: Erlbaum.
Kasari, D., Sigman, M., Mundy, P., & Yirmiya, N. (1992). Affective sharing in the context of joint attention interactions of normal, autistic, and mentally-retarded children.
Kisilevsky, B., Hains, S., Lee, K., Muir, D., Xu, F., Fu, G., et al. (1998). The still-face effect in Chinese and Canadian 3- to- 6-month-old infants. Developmental Psychology, 34, 629–639.
LeDoux, J. E., & Phelps, E. A. (2000). Emotion networks in the brain. In M. Lewis & J. M. Haviland-Jones (Eds.), Handbook of Emotions (2nd ed., pp. 157–172). New York: Guilford.
Lewis, M. (2000). The self-conscious emotions: Embarrassment, shame, pride and guilt. In M. Lewis & J. M. Haviland-Jones (Eds.), The handbook of emotions (2nd ed., pp. 623–636). New York: Guilford.
Lewis, M., Alessandri, S., & Sullivan, M.W. (1992). Differences in shame and pride as a function of children's gender and task difficulty. Child Development, 63, 630–638.
Lewis, M., & Brooks-Gunn, J. (1979). Social cognition and the acquisition of self. New York: Plenum.
Lewis, M., & Michalson, L. (1983). Children's emotions and moods: Developmental theory and measurement. New York: Plenum.
Lewis, M., Ramsay, D. S., & Kawakami, K. (1993). Differences between Japanese infants and Caucasian American infants in behavioral and cortisol response to inoculation. Child Development, 64, 1722–1731.
Lewis, M., Sullivan, M. W., & Alessandri, S. M. (1990). Violation of expectancy and frustration in early infancy: The effect of loss of control. Developmental Psychology, 26(5), 744–753.
Lewis, M., Sullivan, M., & Brooks-Gunn, J. (1985). Emotional behaviour during the learning of a contingency in early infancy. British Journal of Developmental Psychology, 3, 307–316.
Lewis, M., Sullivan, M. W., & Michalson, L. (1984). The cognitive emotional fugue. In C. Izard, J. Kagen, & R. Zajonc (Eds.), Emotions, cognition, and behavior (pp. 264–288). New York: Cambridge University Press.
Lewis, M., Sullivan, M. W., Stanger, C. & Weiss, M. (1989). Self-development and self-conscious emotions. Child Development, 60, 146–156.
Lewis, M., & Thomas, D. (1990). Cortisol release in infants in response to inoculation. Child Development, 61, 50–59.
Lilley, C. M., Craig, K., & Grunau, R. E. (1997). The expression of pain in infants and toddlers: Developmental changes in facial action. Pain, 72, 161–170.
Maikler, V. E. (1991). Effects of a skin refrigerant/anesthetic and age on the pain responses of infants receiving immunizations. Research in Nursing and Health, 14, 397–403.
Malatesta, C. Z., Culver, C., Tesman, J. R., & Shepard, B. (1989). The development of emotion expression during the first two years of life. Monographs of the Society for Research in Child Development, 54 (Serial Nos. 1–2).
Malatesta, C. Z., & Haviland, J. M. (1982). Learning display rules: The socialization of affect expression in infancy. Child Development, 53, 991–1003.
McCall, R. B. (1972). Smiling and vocalization in infants as indices of perceptual-cognitive progress. Merrill-Palmer, 18, 341–347.
Mundy, P., & Willoughby, J. (1996). Nonverbal communication, joint attention and early socioemotional development. In M. Lewis & M. W. Sullivan (Eds.), Emotional development in atypical children (pp. 65–88). Mahwah, NJ: Erlbaum.
Oberlander, T. F. (2001). Pain assessment and management in infants and young children. Infants and Young Children, 14, 33–47.
Oberlander, T. F., Gilbert, G., Chambers, C. T., O'Donnell, M. E., & Craig, K. (1999). Biobehavioral responses to acute pain in adolescents with a significant neurological impairment. The Clinical Journal of Pain, 15, 201–209.
Oster, H. (1978). Facial expression and affect development. In M. Lewis & L. Rosenblum (Eds.), The development of affect (pp. 43–75). New York: Plenum.
Oster, H., Hegley, D., & Nagel, L. (1992). Adult judgements and fine-grained analysis of infant facial expressions: Testing the validity of a priori coding formulas. Developmental Psychology, 28, 1115–1131.
Owens, M., & Todt, E. H. (1985). Pain in infancy: Neonatal reaction to a heel lance. Pain, 20, 77–86.
Papousek, M. (1989). Determinants of responsiveness to infant vocal expression of emotional state. Infant behavior and development, 12, 507–524.
Peters, G., & Czapinski, J. (1990). Positive-negative asymmetry in evaluations: The distinction between affective and informational negativity effects. European Review of Psychology, 1, 33–60.
Preuschoft, S., & van Hoof, Jan A. R. A. (1997). The social function of “smile” and “laughter”: Variations across primate species and societies. In U. Segerstrale & P. Monar (Eds.), Nonverbal communication: Where nature meets culture (pp. 171–190). Mahwah, NJ: Erlbaum.
Ramsay, D. S., & Lewis, M. (1994). Developmental change in infant cortisol and behavioral response to inoculation. Child Development, 65, 1491–1502.
Rosenstein, D., & Oster, H. (1988). Differential response to four basic tastes in newborns. Child Development, 59, 1555–1568.
Rozen, P., Hadt, J., & McCauley, C. (2000). Disgust. In M. Lewis & J. M. Haviland-Jones (Eds.), The handbook of emotions (2nd ed., pp. 637–653). New York: Guilford.
Scanlon-Jones, S., Raag T., & Collins, K. (1990). Smiling in older infants: Form and maternal response. Infant Behavior and Development, 13, 147–165.
Sigman, M., Kasari, D., Kwon, J., & Yirmiya, N. (1992). Responses to the negative emotions of others by autistic, mentally-retarded, and normal children. Child Development, 63, 796–807.
Sroufe, L., & Waters, E. (1976). The ontogenesis of smiling and laughter: A perspective on the organization of development in infancy. Psychological Review, 83, 173–189.
Sroufe, L., & Wunsch, J. P. (1972). The development of laughter in the first year of life. Child Development, 43, 1326–1344.
Steiner, J. E. (1979). Human facial expressions in response to taste and smell stimulation. Advances in Child Development and Behavior, 13, 257–296.
Stenberg, C. R., Campos, J., & Emde, R. (1983). The facial expression of anger in seven-month-old infants. Child Development, 54, 178–184.
Sullivan, M. W., & Lewis, M. (1988). Facial expressions during learning in 1-year-old infants. Infant Behavior and Development, 11, 369–373.
Sullivan, M., & Lewis, M. (1989). Emotion and cognition in infancy: Facial expressions during contingency learning. International Journal of Behavioral Development, 12(2), 221–237.
Sullivan, M.W., & Lewis, M. (1990). Contingency Intervention: A program portrait. Journal of Early Intervention, 14(4), 367–375.
Sullivan, M. W., & Lewis, M. (in press). Contextual determinants of infant anger. Developmental Psychology.
Sullivan, M. W., Lewis, M., & Alessandri, M. (1992). Cross-age stability in emotional expressions during learning and extinction. Developmental Psychology, 28, 58–63.
Sullivan, M. W., Ramsay, D., & Lewis, M. (1992). Learning and temperament in young infants: A longitudinal study of 2- to 8-month-olds. A paper presented at the Biennial Meeting of the International Society for Infant Studies, Miami, FL.
Vine, I. (1973). Facial visual signalling and social development. In M. von Cranash & I. Vine (Eds.), Social communication (pp. 195–289). New York: Academic Press.
Walden, T., & Knieps, L. (1996). Reading and responding to social signals. In M. Lewis & M. W. Sullivan (Eds.), Emotional development in atypical children (pp. 29–42). Mahwah, NJ: Erlbaum.
Watson, J. B., & Rayner, R. (1920). Conditioned emotional reactions. Journal of Experimental Psychology, 3, 1–14.
Yirmiya, N., Kasari, C., Sigman, M., & Mundy, P. (1989). Facial expressions of affect in autistic, mentally retarded, and normal children. Journal of Child Psychology and Psychiatry, 30, 725–735.
Zajonc, R. B. (2001). Mere exposure: A gateway to the subliminal. Current Directions in Psychological Science, 10, 224–228.

emotion; emotional development; facial expressions; infants; nonverbal communication

©2003Lippincott Williams & Wilkins, Inc.