Motivation and Emotion, VoL 5, No. 1, 1981

Infant E m o t i o n and the Vocal A f f e c t Lexicon 1 Carol Zander Malatesta 2 Rutgers, The State University

The role o f the vocal channel o f emotion expression in infancy has been neglected in developmental theory. The present review describes the ontogenetic course of vocal emotional expression as exhibited by human and infrahuman primate young and considers its dynamic relationship to the facial and bodily components of expression. The infant's encoding of negative and positive emotion expression is discussed within a developmental framework. In addition, this review assesses the impact of early social influences. It is concluded that early patterns o f infant vocal emotional expression are probably biogenetically determined and that there may be certain universal vocal signals. However, data derived from studies o f dyadic interaction indicate that the transition from raw affect expression in early infancy to a more modulated pattern later on is a product not only of neuromuscular maturation but of maternal coaching in affective expression as well.

In 1966 Peter Wolff noted that there were no adequate theories of affective development. More recently, Lewis (1979) suggested that the problem still exists. The lack of progress in this area is partially explained by the rather slow accumulation of basic descriptive data. In addition, existing models of affective development have relied almost exclusively on data detailing the ontogenesis of facial expressions. While research on facial expressions has done much to advance the scientific study of the emotions, any theory that

'The author thanks David Brodzinsky, Richard Curtin, and Horst D. Steklis for their critical reading of the manuscript. Special thanks are also due Rae Carlson, Jeannette Haviland, and Carroll Izard for earlier suggestions. 2Address all correspondence to Carol Zander Malatesta, Livingston College Psychology Department, Rutgers University, New Brunswick, New Jersey 08903. I 0146-7239/81/0300-0001503.00/0 © 1981 Plenum Publishing Corooration

2

Malatesta

hopes to do justice to affects developmentally must not neglect the other aspects of expressive behavior. Affect expression should be conceived of as a tripartite complex of somatic activity involving vocal and bodily (postural/gestural) dements as well as facial behaviors. Although there has been little research on the bodily expression of emotion in infancy, there is a growing amount of literature on the vocal component. These data, however, have yet to be treated systematically and examined within a developmental framework. It is the purpose of the present paper to provide such a framework. The first part of this paper discusses vocalization as affective communication and provides a description of the ontogenetic course of vocalization in the nonhuman primate infant. The following section considers the human infant's encoding of positive and negative emotion both vocally and facially. Finally, the impact of early social influences in modulating early infant affective expression is considered.

VOCALIZATION AS AFFECTIVE COMMUNICATION Before approaching the human developmental literature, it may be instructive to examine some of the properties of affective communication in nonhuman primates. Both Izard (1977) and Ekman (1973) have discussed the evidence suggesting that human facial expressions have evolved from nonhuman primate facial displays and find that the evidence is strongly in favor of phyletic continuity. Scherer (1979) suggests that the same may be true for the vocal expression of affect. In fact, the literature to be reviewed in this paper would seem to bear this out. The first part of the present section reviews some common properties of the affect communication system as a whole in mature members of the various primate species. This is then followed by a more detailed discussion of primate ontogenesis with special attention to the development of the vocal expression of affect.

Common Properties of the Affect Communication System in Mature Members of Various Primate Species Adult primates possess an effective communication system based on the signaling of affective states. The following features are common properties of the affect communication system. Redundancy in Affect Expression. Because affect may be encoded in several different channels, either concurrently or sequentially, there is inherent redundancy in signal information (Chevalier-Skolnikoff, 1973; Andrew, 1963; Green, 1975).

Infant Emotion

3

Variable Temporal Sequencing of the Elements of Affect Expression. Emotional elements--behaviors from different body parts and different sensory modes--form affective constellations; some patterns or groupings of elements usually occur together, while others vary independently (Chevalier-Skolnikoff, 1973; Marler, 1965). Despite the fact that vocal and facial expressions have apparently evolved together (Andrew, 1963), some initial evidence indicates that there may be phylogenetic progression in the degree to which vocal and facial expressions of affect can be exercised independently, with the greatest degree of possible dissociation evidenced in man. (See charts provided by Chevalier-Skolnikoff, 1973, pp. 66-79.) Intergradation of Signals. "Graded" signals differ from "discrete" signals in that the latter are generally thought of as relatively nonoverlapping classes of behavior (Takeda, 1965); graded signals, on the other hand, blend, are variable, and increase in intensity with the greater motivation of the signaler (Murray, 1979). Gradedness of signals within an affect class and intergradations between signals of different classes are thought to constitute an "analogue qualifier system" (Scherer, 1979), providing for various shades of meaning in communication (Marter, 1965; Green, 1975). Gradedness of signals accompanies progression up the phylogenetic scale, with gradedness of facial as well as vocal signals a striking feature in the Old World monkeys and apes, especially among those that are ground-dwelling (Darwin, 1872/1965; Redican, 1975; Green, 1975; Chevalier-Skolnikoff, 1973; Andrew, 1963). Adult humans are said to show almost continuous gradations among facial expressions (Bastian, 1965), and there is apparently significant gradation in the vocal indicators of affect in the paralinguistic aspects of human speech (Scherer, 1979). In summary, then, the adult primate possesses an effective (redundant) communication system and two means by which to convey complex and subtle messages, i.e., by grading and blending affects within any particular channel and also by the temporal sequencing of the three channels of expression. The existence of three channels and two processes that may modify the signal value of any single message transmission guarantees a richness of information; at the same time, the phenomenon of redundancy ensures that this surfeit of nuance is not bought at the expense of intelligibility. Obviously, this is a tremendously intricate system, and the developmentalist may well ask how the formal and functional aspects of such a complex system are rendered ontogenetically.

Primate Ontogenesis and the VocalExpression of Affect A search of the primate literature for developmental data proved to be less illuminating than anticipated. Most accounts of the facial expressions of nonhuman primates dwell on the expressions of adult members of the

4

Malatesta

species. Observations of the ontogeny of facial expressions seem particularly difficult to obtain, apparently because neonates display few facial expressions, at least as observed in captivity and seminatural living conditions (Rowell, 1963; Hinde, Rowell, & Spencer-Booth, 1964; Brandt, Baysinger, & Mitchell, 1962). On the other hand, most of these same primate infants have a number and variety of vocal expressions that are exercised readily and volubly; the relatively low threshold for vocalization and relative poverty of facial expressions early in infancy gives way to the more predominant use of the facial-visual communicative system later on (Redican, 1975). Before turning to the human literature I will summarize some of the more pertinent aspects of the vocal expression of affect in one species of Old World monkey, Macaca fuscata, the Japanese macaque; to this end I rely on two studies by Takeda (1965, 1966). The work of Takeda possibly represents the most thoroughgoing and detailed scientific document available on the ontogenetic course of affect vocalization in primates; it is remarkable not only for its abundant detail concerning vocal behavior but also in its careful documentation of other ongoing developmental phenomena--data that are indispensable when it comes to interpreting the semantics of primate vocalization. These data indicate that innate (biogenetic and maturational) factors control the early expression of vocal affect but that learned factors may come into play later on, leading to the voluntary control of at least some vocalization. Takeda speaks of six original vocal stock sounds from which all later vocal behavior derives. A vocal stock is defined as a lineage of vocal sounds and the process of their functional progress in the monkey's life history. Sounds are classified as "derivatives" on the basis of acoustic properties as well as functional similarity of behaviors that accompany them. The six stock sounds are designated by the alphabetic letters A through F. All the vocal sounds ranging from the A to the E stock are connected with emotions of displeasure, dissatisfaction, pain, or sense of danger. The F stock sounds, which proliferate later developmentally, convey mainly positive exicitement and "nonemotional," declamatory expressions, the latter are remarkable for their variety and plasticity. No specific behaviors appear to be linked with the emission of these sounds. Developmentally there is differentiation and then elaboration and accentuation of the C and B stock sounds: During the first 3 weeks of life the young Japanese monkey already possesses a group of crying sounds (C stock); these are high-pitched, intense sounds of long duration, and tones fluctuate with shifting emotional states. Vocalizations of sullenness, sulking anger, and finally what are apparently temper tantrums appear during the next two stages (4-8 and 8-15 weeks); these derive from the early C stock sounds and as such there is continuity between the early cry and later anger vocalizations, a point to keep in mind when reviewing the human cry literature. Crying and high-pitched screams are often accompanied by

Infant Emotion

5

defensive facial expressions, such as drawing back the corners of the mouth to display the teeth and gums. Sobbing sounds are sometimes accompanied by pouting of the mouth. During the second stage (4-8 weeks) the baby begins its forays away from the mother in active exploration, and evidence of curiosity and fear are first seen. This stage ushers in fear sounds, which are derivatives from the B stock sounds that earlier were given in response to anxiety about loss of balance and the like, in association with mild tension. The fear sounds are more intense acoustically than the earlier tension sounds, and tend to be accompanied by defensive expressions. F stock sounds are the "calm" sounds and those accompanying light excitement; they proliferate during the third stage (7-15 weeks) and may or may not be accompanied by facial expressions; the calm, more "neutral"sounding ones tend to lack facial expression, while those expressing excitement and greeting (joy) tend to be accompanied by vivid facial expressions and movements. During the fourth stage (16-30 weeks) no new emotional sounds are introduced aside from some intense supplicating sounds. There is mainly elaboration of F sounds; these sounds are generally low in tone and do not involve violent emotional activities. Rather they are emitted by slight motivations (mild tension, excitement, nonurgent demands, greetings). These sounds are referred to as "mutterings" or "conversational sounds"; they are used in short-distance communication and appear in the monkey's repertoire as sounds of reply to other animals. In the adult macaque the calmer sounds continue to be used in one-to-one communication (Itani, 1963). Emotional sounds (especially those of the strong emotions such as fear) are used for one-to-many communication and are more likely to be accompanied by facial and postural movements. The affects sounds are said to "strengthen" the emotional meaning of a display; the vocal expression of strong emotion can generate excitement among troop members, and its usefulness in consolidating the troop for tight or flight is apparent. To summarize, there are several noteworthy characteristics about affect vocalization in the Japanese monkey. 1. Displeasure sounds, which emerge first and are more diverse than pleasure sounds early in development, diminish in frequency with the onset and proliferation of conversational sounds, suggesting modulation of affect expression. The displeasure sounds are those connected with the more urgent survival needs, and their reflexive quality, lower threshold of activation, and early emergence in development reflect their adaptive nature. 2. A general property of affective vocalization, especially the intense kind preserved in adulthood, is that it acts as a general motivational amplitier, boosting the gain of affect signals traveling by other channels. 3. Joy/interest affect vocalization and conversational vocalizations appear to form one suprasegmental class of vocalizations while the remainder of affect vocalizations form another class.

6

Malatesta

4. The progression from discrete calls early in infancy to "differentiated" sounds and the graded "conversational" subtlety found later in development suggests some degree of learned control of motivational-communicative acts. The above observations indicate that both innate and learned factors play a part in the ontogenesis of the vocal expression of affect in the Japanese macaque. Fragmentary evidence from other sources indicates that the above patterns are not unique to Macacafuscata but may be typical of most nonhuman primates (Steklis & Raleigh, 1979). We may now ask whether or not these same patterns obtain for Homo sapiens. To answer this question we turn to the human developmental literature. As well be seen, there are a great deal more detailed data available on the vocal expression of affect in human infants than in nonhuman primates. In the following section, which discusses the human literature, the ontogenesis of negative vocal expressions and positive vocal expressions will be treated separately.

EMOTIONAL ONTOGENESIS IN THE HUMAN INFANT

The Infant's Encoding of Negative Affect During the 1st year of life, the vocal manifestation of negative affect is almost exclusively encoded by crying. Within the human developmental literature there has been a good deal of controversy over whether crying encodes differential affect states. With respect to this issue there are essentially two schools of thought. One school, represented by investigators such as Murray (1979) and Zeskind and Lester (1978), regard all crying as an expression of an undifferentiated state of distress or discomfort, although they do allow the possibility that gradations in crying may code degree of distress or discomfort. The reasons for adopting this stance are apparently both theoretical and empirical. Murray (1979) has suggested, for example, that humans evolved as a carrying species whereby infants were carried and in continuous contact with their mothers. Mothers could judge infant precry cues and minister to the child's needs before crying began in earnest, and thus there would be no need for differential vocal cue specificity to evolve. Proponents of the "undifferentiated cry" school also frequently invoke references to studies that have failed to find evidence for the reliable discrimination of cries elicited under different conditions (Sherman, 1927, 1928; Muller, Hollien, & Murray, 1974; Aldrich, Sung, & Knopp, 1945).

Infant Emotion

7

A second school of thought, in this case represented by writers such as Wolff (1965), Tomkins (1963), and Darwin (1877), suggest that there are or may be different types of cries that either are determined innately or become differentiated over time. Proponents of this school cite their own sources of confirmation (e.g., Valanne, Vuorenkoski, Partanen, Lind, & WaszHockert, 1967; Wasz-Hockert, Lind, Vuorenkoski, Partanen, & Valanne, 1968). There are, in fact, a whole host of empirical typological studies, some of which are based on mechanical analysis in the form of sound spectrography and others employing human decoders such as nurses, psychology students, and mothers. The sound spectrography studies will be reviewed first.

Cry TypologyStudies MechanicalAnalysis. There are three studies that fall into this class, those of Wasz-Hockert et al. (1968), Wolff (1965), and Murray, Amundson, & Hollien (1978). Wasz-Hockert et al. recorded cries elicited under conditions of birth, pain, hunger, and pleasure for spectrographic analysis. According to this report, the four types of cries could be discriminated on the basis of distinctive acoustic parameters. For example, the pain cry is said to be long and characterized by a high maximum pitch, a falling melody, subharmonic breaks, and vocal fry, whereas the hunger cry is said to have a characteristic rising-falling melody and lower maximum pitch with few subharmonic breaks. While Wasz-Hockert et al. did find differences discriminating the various cries, they did not find developmental changes in cry patterns; this latter result is perhaps due to the insensitive nature of their cross-sectional format, since other researchers using a semilongitudinal design did find developmental changes with respect to respiratory parameters during crying (Wilder & Baken, 1978). Wolff (1965) also studied the cries of young infants (1-6 months old) and has asserted distinctive acoustic and spectrographic features of hunger, " m a d , " pain, and frustration cries. However, it is important to note that characterization of the different cries in both the Wasz-Hockert et al. and Wolff studies are based on subjective classification schemes despite the use of spectrography; the parameters of the cries were not described numerically (except for fundamental frequency in some cases) or subjected to statistical analysis. Murray et al. (1978), however, did employ statistical analysis. These researchers examined the fundamental frequency (F6) of the cries of infants 3 - 6 days of age in response to pain, hunger, and startle stimuli. They found no significant difference in the mean F6 among the cry types. Despite this, it

8

Malatesta

cannot be concluded that there are no measureable and statistically significant differences acoustically, or that the cries are not phenomenologically different for the infant or caregiver. While the overall rank-order FO's in increasing value were pain, hunger, and startle, individual differences in rank order were remarkable. Whether these differences between subjects are due to real variability in the Fgt of cries without respect to elieiting condition, or whether individual difference patterns are reliable and stable within individuals and therefore discriminable to someone well acquainted with the patterns, probably cannot be ascertained by this design, which was purely cross-sectional. The application of cross-sectional data to a question that can perhaps only be answered by a longitudinal or repeated-measures design may lead to erroneous conclusions. In point of fact, an early study by Fairbanks (1942), in which the hunger cries of one infant were examined longitudinally during the first 4 months of life, found that there was a progressive increase in F6, which the author ascribed to variation in vocal fold tension. In addition, Ringel & Kluppel (1964) have reported wide individual differences among babies in terms of cry characteristics although patterns seemed to be stable for each infant. Thus, perhaps one reason mothers profess to recognize differences in their own infant's cries is that they become sensitized to their infant's unique pattern of response. Furthermore, it is doubtful that discrimiantion of infant crying is based on so simple a property as F6; it is more likely that mothers rely on pattern discrimination, i.e., the onset, offset,'and stayset properties of the cry, and to harmonics (timber) and other subtleties. Human analysis. Typological studies employing a cry discrimination paradigm, in which subjects listen to and make judgments as to the probable cause of a particular cry or its affective meaning, have a longer history than those using sound spectrography, the first being conducted in 1927 by Mandel Sherman. A summary of her results and those of other subsequent studies is presented in Table I. It can be seen that three of the studies that were unsupportive of discrimination employed very young infants. As such it is not surprising that they failed to find evidence of discriminability. The immature status of the vocal apparatus and central nervous system mechanisms is such that stable patterns of crying are highly unlikely. Lack of experience in the form of auditory feedback to the self and contingent responding by caregivers may be other factors working against differentiation and discriminability. Lenneberg, Rebelsky, and Nichols (1965), in a study of the infants of normally hearing and deaf parents, in fact, found a wider range of crying behavior among the neonates of the deaf, presumably attributable to the less contingent responding of the deaf parents. Normally, crying patterns do appear to become more stable and less variable over time (Prechtl, Theorell, Gramsbergen, & Lind, 1969).

Infant Emotion

9

Table I. Cry Discrimination Studies i

Study

Infant age

Sherman, 1927

3 - 7 days

Aldrich, Sung, and Knopp, 1945

1-8 days

Wasz-Hockert, Partanen, Vuorenkoski, Michelsson, and Valanne, 1964

Unspecified mean age (range: 0 - 8 mo.)

Wasz-Hockert, Partanen, Vuorenkoski, Valanne, and Michelsson, 1964

Unspecified mean age (range: 0 - 8 moo

Valanne, Vuorenkoski, Partanen, Lind, and Wasz-Hockert, 1967

1-7 days

Wasz-Hockert, Lind, Vuorenkoski, Partanen, and Valanne, 1968

Unspecified mean age (range: 0 - 7 too.)

Muller, Hollien, and Murry, 1974 Wiesenfeld, Malatesta, and DeLoach, 1981

Unspecified mean age (range: 3-5 mo.) Mean age: 154 days

Findings Subjects unable to associate the elicited cries with the stimulus situation Unable to obtain clear-cut answers since the subjects reported that more cries were due to "unknown causes" than to any other stimuli Trained nurses without children of own but some having had experience with children identified 67% of cry types (above chance) Cries discriminated; best performance by childexperienced men and women Asked to pick the two hunger cries of own infant from those of other, about 1/3 did better than chance Cries discriminated; best performance by midwives, children's nurses, and other experienced women Unsupportive of discrimination Significant cry discrimination accuracy by mothers hearing own infant's cries

T h e f o r e g o i n g facts m a y h e l p e x p l a i n t h e d i f f e r e n t i a l results o f c r y d i s c r i m i n a t i o n studies. I f crying b e c o m e s m o r e s t a b l e d e v e l o p m e n t a l l y a n d c h a r a c t e r i s t i c p a t t e r n s g r a d u a l l y evolve, o n e w o u l d a n t i c i p a t e m o r e p o s i t i v e results in c r y d i s c r i m i n a t i o n studies using o l d e r i n f a n t s . O f t h o s e studies using o l d e r i n f a n t s ( W a s z - H o c k e r t et al., 1968; W i e s e n f e l d , M a l a t e s t a , & D e L o a c h , 1981; W a s z - H o c k e r t , P a r t a n e n , V u o r e n k o s k i , M i c h e l s s o n , & V a l a n n e , 1964; W a s z - H o c k e r t , P a r t a n e n , V u o r e n k o s k i , V a l a n n e , & M i c h e l s s o n , 1964; M u l l e r et al., 1974), f o u r o u t o f five f o u n d evidence o f d i s c r i m i n a b i l i t y , a n d t h e d i r e c t i o n o f f i n d i n g s indicates t h a t s u b j e c t experience with i n f a n t s is a d e c i d e d f a c t o r in j u d g m e n t a c c u r a c y . T h e

10

Malatesta

surprising thing about the Muller et al. study, the only one failing to find evidence of discriminability, is not so much that their unrelated adults could not decode the cries with a high degree of reliability but that their subsample of mothers listerning to cries of their own infants did no better than unrelated mothers. I~fferential methodology for cry elicitation and playback may have contributed to the discrepant findings. Another possible source is difference in infant age--although the infants were said to range in a g e from 3 to 5 months, the mean age was not reported and may have been on the low end of the range. There are apparently no studies of infant cries employing infants older than a mean age of 5 months. This is unfortunate since it is certainly possible that the affective quality of crying undergoes some degree of differentiation after the early months, as in the case of nonhuman primates (Takeda, 1965, 1966). Gross motor patterns and facial expressions apparently differentially encode anger and distress states in the older infant. Izard (personal communication) maintains that anger and pain cries of infants are accompanied by differential facial configurations, especially about the yes--in the anger cry the eyes are fixed and staring, whereas the initial response to unanticipated pain is one of scrouged eye fissures. The motoric pattern and tension of the body is said to be different under conditions of anger, temper tantrums, and distress crying in both humans (Bridges, 1932) and primates (Andrew, 1963; Takeda, 1965, 1966). In chimps, for example, during distress crying the animal becomes more passive and unresponsive as vocalizations become less intense, but when the animal is deprived of a desired object, locomotor responses and aggressive responses are facilitated and there is intense vocalization. In light of the fact that distinctive vocal patterns tend to accompany distinctive facial and motoric patterns during intense motivational states in lower primates (Chivalier-Skolnikoff, 1973) and that differential facial and bodily responses are seen in painful and frustrating situations with human infants, there is every reason to believe that a distinctive anger cry exists. There may also be a "fear cry" with distinctive acoustic properties. During the approach of a stranger at about 7 months, Bridges (1932) noted, an infant may cry; during the cry the body is said to be held rigid and inactive, much in contrast to the wild flailing and thrashing of the body in states of obvious anger. That the cry may in fact sound different acoustically is suggested not only by the apparent differential tension distribution in the body under conditions thought to evoke anger versus fear but also on the basis of the fact that stranger anxiety is a newly emergent phenomenon, regarded by some as a manifestation of a new psychic (affective?) organizer (Spitz, 1965), or as representing a shift to a new level of central nervous system organization (Emde, Gaensbauer, & Harmon, 1976). In addition, the precry face described for the fear response (Emde et al., 1976) appears

Infant Emotion

11

to be different from that described as preceding the neonate distress face (Oster, 1978). The primate literature also suggests that this is a period that may herald a change in patterns of affect vocalization. Recall that as the infant macaque begins to move away from its mother to explore the world independently, it first begins to manifest a full-fledged fear response that includes a new fear-distress vocalization (Takeda, 1965, 1966; Bronson, 1968). Further study is warranted. Unlike Murray (1979), who views typological studies as representing an "overintellectualization of the problem" (p. 198), I regard further work in this area as potentially useful in delineating the process of affective development and reflecting underlying neural mechanisms and levels of neural organization. Almost all studies of the infant cry, whether they be topological studies of the sort described above using normal children or attempts to relate acoustic patterns of the cry to brain damage, have explicitly or implicitly assumed a relationship between phonatory patterns and the infant's neurophysiological status. A recent study by Wilder and Baken (1978) speaks to this issue. Their research on patterns of crying suggests that central nervous system maturation during the 1st year is an uneven process, possibly engendering stagelike changes in the experience or expression of the affect carried by the cry. In their semilongitudinal study of cry respiratory patterns, these researchers have demonstrated distinct changes in the temporal aspects of respiration over 8 months. The infant apparently acquires the ability to control in a saltatory fashion the complex motor coordination required for the negotiation of rapid inspiration and prolonged, mostly voiced, expiration; there are salient changes taking place between the 1st and 2nd month, the 5th and 6th month, and the 7th and 8th month. These authors view their findings as reflecting growth in neuromotor control; to the extent that affect is a neuromotor event as well, these data provide interesting food for thought as to whether one can expect to find evidence of changes in affect expression during these same epochs. Interestingly, several authors have placed the emergence of fear at around the 6th month (see Izard & Buechler, 1978, for a review), and the emergence of fear in the form of stranger anxiety is said to occur at about 8 months (Bridges, 1932; Spitz, 1965).

Summary In this section we considered the issue of whether or not crying encodes specific affective states in the young infant. Although more work is needed in this area before a definitive answer is forthcoming, the evidence thus far indicates that crying may indeed designate differential motivational or affective states. Cries elicited under different conditions appear to be

12

Malatesta

characterized by distinctive acoustic properties, and differential conditions can be detected by at least some subjects on the basis of auditory cues alone, especially in older infants. (Some period of matruration or learning may be necessary for the emergence of certain affective states such as anger or fear.) Evidence of constellations of distinctive facial and gross motor patterns under differential eliciting conditions further bolsters the argument that there may be different types of cries encoding differential states.

THE INFANT'S ENCODING OF POSITIVE AFFECT While it is common to attribute unpleasant affective experience to the crying infant and positive affect to infants engaged in smiling, laughing, and cooing, there is perhaps greater need to exercise caution in inferring the phenomenology of the infant under the latter conditions than under the condition of crying. Nevertheless, if for the moment we disregard all the usual caveats against this type of adultomorphic inference and accord at least some positive valence to these expressions, we may then proceed to ask questions such as: When in ontogenesis is positive affect first encoded and what form of expression does it take developmentally?

Smiling and Vocalization In this section we will consider smiling and cooing together, since both types of behaviors are said to index positive emotional states and since they may in fact form a constellation of expressive behavior. First of all, however, we need to acknowledge that there is some question as to when smiling itself may be regarded as reflecting a positive emotional state. Smiles first occur during REM sleep very early in life. Neonatal REM similes are fleeting, poorly articulated, and appear to be reflexive responses to random endogenous stimulation (Emde et al., 1976); few would attribute positive emotional experience to the infant in this state. Smiling in response to the human face during the awake state may be seen during the 1st month; at the same time, it may also be given in response to a simple pattern of dots or angles or to an inanimate object suddenly appearing in the visual field (Ahrens, 1954; Wolff, 1966). As such, the affect encoded might in fact be interest or surprise rather than pleasure. We must at least entertain the possibility that early smile behavior, unaccompanied by positively toned vocalization, may be mainly reflexive behavior or random movement associated with any number of underlying affective experiences. It would be helpful to have a concurrent index of affect state, and vocal expression is at least one potential candidate. The relationship between smiling and vocali-

Infant Emotion

13

zation, however, has apparently never been explored in any systematic way, and objective measures of the tonal quality of vocalizations are just beginning to be developed (cf. Young & Decarie, 1977). Meanwhile, we simply note that while there are copious studies of smiling, there are but few that describe vocalization that occurs proximate to smiling, and fewer that reference the tonal quality of vocalizations. The following section summarizes the existing literature on smiling and vocalization. Assessment of the developmental relationship between smiling and vocalization is made difficult by the presence in the literature of different methods of describing infant sounds. The picture is complicated further by the lack of satisfactory system of classification. Thus it is not surprising that so little is known about the ontogeny of infant vocalization, not to mention the relationship between smiling and vocalization. A search of the literature turned up only five studies that provide a significant degree of information regarding the temporal relationship between smiling and cooing--those of Lennberg et al. (1965), Washburn (1929), Stark, Rose, and McLagen (1975), Stark (1978), and Young and Decarie (1977). As such, the studies will be treated separately and their findings described in some detail. Lenneberg et al. (1965) recorded the vocalizations of two groups of infants during the first 3 months of life--those born to deaf parents and those born to hearing parents. They describe" cooing" vocalizations, which are defined as "happy" sounds, having no strained quality, high in vocality, low in glottal stops, and having a smooth onset. There were no differences in age at onset or frequency of occurrence in the two groups of infants studied. Sounds classed as cooing, according to their findings, first emerge during the 2nd month and are at their height by the end of the 3rd month. They are reported as regularly accompanying the smiling response and "are thus closely tied to the stimuli eliciting smiling." According to the authors, cooing behavior emerges quite distinctly from crying, and there is no gradual transition from one to the other. (This position, it will be seen, is in marked contrast to that of Wolff, and Stark and colleagues.) Other authors have noted that cooing becomes more variegated over the ensuing months in normal children, and by about 6 months there is evidence of adult intonational patterns in their babbling (Kaplan & Kaplan, 1971). One of the best ontogenetic treatments of smiling is to be found in Washburn's (1929) impressively detailed monograph on the smiling and laughter of 15 infants observed in a combined longitudinal cross-sectional design. Although there was apparently no systematic recording of the cooccurrence of smiling and vocalization, the author does note that early smiles were "very often accompanied by vocalization," a finding consistent with that of Lenneberg et al. Smiling with vocalization was most typical at 8, 12, and 16 weeks. By 36 weeks the variety and number of syllables employed increased with smiling, and almost any of the sounds in the infant's

14

Malatesta

increasingly expressive jargon appeared to be coupled with smiling. Washburn also reports on motor activity accompanying smiling and laughter, the only author in this group to have done so. Generalized activity of the arms and legs, rather like kicking or waving, is described as coming just at the end of the smile or laugh; the presmiling or prelaughing period was generally one of mobilized attention with stilled activity. Washburn's data, then, indicate that the expression of positive affect engenders a constellation of somatic activity involving all three channels of affect expression. According to the data she furnishes, this pattern obtains from the onset of smiling at 2 months through the 1st year of life. Brazelton, Koslowski, and Main (1974) have also reported on the cycling of the arms and legs that may occur in conjuction with smiling or vocalization. Interestingly, the gross motor activity is said to be more intense with vocalization, suggesting that the feedback from the vocal expression of affect may serve to recruit more intense affective involvement. This finding is evocative or Tomkins's notion of the "self-reproducing" quality of affect (1962, 1963). Since the vocal affect system is said to be phylogenetically older than the facial affect system (Andrew, 1963), it is not surprising that it may be more affectively arousing. In 1965 Peter Wolff published the results of his longitudinal study of the vocal output of 18 infants. In this paper he advanced the notion that the earliest noncry vocalizations are functionally and morphologically related to crying; in emphasizing that new noncry patterns of vocalization seem to appear for the first time as the infant gets fussy and shortly before he starts to cry, Wolff comes close to suggesting that discomfort and comfort sounds may derive from the same vocal stock, a notion that finds explicit expression in two recent papers by Stark and colleagues (Stark, 1978; Stark et al., 1975) and that is based on their own longitudinal study of the discomfort, comfort, and vegetative vocalizations of two female infants. However, it is not at all clear from either the Wolff or Stark et al. studies that the acoustic similarity between discomfort sounds and those occurring in more contented states is sufficient to infer the existence of common functional or affective forerunners. With reference to the Stark and colleagues' notion of one common vocal stock, it should be noted that this is apparently not the developmental course in our more advanced primate relatives; pleasue and discomfort vocal affect expressions, as well as "neutral" or "conversational" mutterings, evolve from separate vocal stocks (Itani, 1963; Tekeda, 1965, 1966). The two studies by Stark and co-workers also leave some room for doubt that they are describing the same vocal phenomena as Washburn: The comfort sounds are described as "cooing" sounds, and some have "non-sonorant consonant-like elements." Washburn, on the other hand, describes " a h ' s , " squeals, and gurglings. There is even further reason to suspect that the studies are not comparable; Washburn suggests a high cooccurrence of smiling and vocalization early on, while Stark (1978) notes

Infant Emotion

15

that the comfort sounds were infrequent at the time of the emergence of smiling, but with a gradual convergence of these two phenomena 11A to 2 weeks later. Another constraint to reliance on the Stark et al. data is their use of only two subjects and only one sex. The above considerations thus preclude judgment as to whether comfort and discomfort sounds do indeed share common developmental ancestors either morphologically, semantically, or affectively. This would seem to be an important area for further research inasmuch as it touches on what may be the developmental predecessaors of the paralinguistic aspects of human speech. "Paralinguistics" are those qualities o f the voice th at convey the emotional components of meaning in communication (Helfrich, 1979). Recently, Scherer (1979) reviewed an array of studies concerning the vocal indicators of affect in adult subjects and concluded that there was sufficient support for the notion not only that paralinguistic features of speech (pitch, intonation, etc.) could encode nonspecific arousal but that the "major dimensions of emotional experience may be differentially expressed by patterns of vocal cues and that discrete emotions are characterized by their position in this dimensional space" (p. 515). Research with infants and young children suggests that there may be an early and innate natural tendency to intonate speech with appropriate emotional expression. On the basis of spectrographic and qualitative analysis, Tonkova-Yampol'skaya (1968) has identified patterns of emotional meaning inhering in the cry and noncry vocalizations of infants during the first 2 years of life that have counterparts in adult intonational patterns. For example, the expressive calm cooing sound of infants is said to resemble the intonation of affirmation in adults, and the intonation of discomfort in adults has its homologue in the contour pattern of distress sounds of earliest infancy. Ricks (1979) tape-recorded sounds of young infants given in response to classes of stimulus events and played them back to parents. The infants were 8 to 12 months old and from a variety of language backgrounds. He found that the sounds made by the various infants were strikingly similar. Four common and easily identifiable sounds were a requesting sound, a frustrated sound, a greeting sound, and a sound expressing pleased surprise at a novel and exciting event. Parents found it easy to understand the message conveyed by the intonated sounds not only of their own children but of others as well. The above studies suggest that the tendency to intonate speech affectively may have an early and innate origin, although additional meaning may be mapped on through learning (Weeks, 1971).

Laughter Depending on criteria of judgment, laughter is said to emerge variously at a median age of 9 weeks (Dennis & Dennis, 1937), at the end of the

16

Malatesta

1st month (Wolff, 1965), at 12 weeks (Washburn, 1929), or at 4 months (Sroufe & Waters, 1976). In any event, laughter is generally regarded as a later developmental phenomenon than smiling. Most writers accept laughter as an intense affective experience, though there is some disagreement as to whether it should be regarded as chiefly the "anticipation" of pleasure (Darwin, 1872/1965) or its actual enjoyment (Charlesworth & Kreutzer, 1973), and whether or not it is a pure affect in the sense of positive affect uncontaminated by negative affect, or whether it is an expression of conflicting emotion (Wolff, 1965; Washburn, 1929). In primates, the "play face," consisting of a relaxed open-mouth display often accompanied by quick and shallow staccato breathing with or without vocalization, is thought to be phylogenetically closely related to laughter in humans (van Hooff, 1969). In one of the rare ontogenetic primate studies conducted in the wild, Plooij (1979) most recently found that the play face emerges at 6 weeks in the young chimpanzee; the "laugh" sound emerges during the 3rd month and is given in response to tickling by the mother. Although the author cautions against interspeciflc comparisons, one cannot help but be impressed by the similarity between the course and timing of emergence of the play face and laughter in the chimp on the one hand, and the human smile and laughter on the other. There are also interesting parallels in the way that the human infant smile and the primate infant's use of the play face draw forth interactive behavior from the mother. While van H o o f (1969) has concluded that the human smile has evolved from the "bared-teeth-face," which is regarded as a submissive gesture in the higher primates, rather than the play face, Plooij's datawould seem to reopen the case. In human infants the laugh is said to accompany several different facial expressions (Young & Decarie, 1977). According to Washburn (1929), the human laugh has a distinct quality and is stereotypic in sound even in the young infant, there being far less variation than with the vocal sounds accompanying smiling. Overall, there is said to be little developmental change in the expression of laughter vocally or motoricaUy at least during the 1st year of life. Research on laughter beyond infancy has focused exclusively on the nature of situations eliciting it (Charlesworth & Kreutzer, 1973; Sroufe & Waters, 1976; Sroufe & Wunsch, 1972), with stereotypy of motoric, vocal, and facial components of the expression apparently taken for granted.

Summary In the foregoing sections we considered the ontogenetic course of smiling, cooing, and laughter. There are several noteworthy points emerging from the literature reviewed. 1. The expression of positive affect quite frequently is a "whole-body" phenomenon, involving vocal, facial, and gross motor activity, as is the case

Infant Emotion

17

of negative affect expression. However, a greater degree of dissociation among the channels of affect expression is possible in positive as compared to negative states. 2. Controversy exists with respect to the issue of whether comfort and discomfort sounds arise from the same or different vocal stocks. Although this issue is far from being resolved, the bulk of evidence would seem to favor the notion of separate vocal stocks, in keeping with the nonhuman primate pattern. 3. Laughter is apparently a rather stereotypic response, at least accoustically. This is not the case with other pleasure sounds. Data on the emergent properties of affective vocalization--the vocal behavior of infants born to deaf and hearing parents--as well as data based on the vocalizations of infants from differing language backgrounds indicate that there may be certain universal vocal signals, as is the case with facial expressions (Ekman, Sorenson, & Friesen, 1969; Ekman & Friesen, 1971; Izard, 1971). However, other data demonstrating the proliferation of more variegated sounds and changes in intonational patterns later on point to the probable contribution of learning during maturation. With respect to the last point, we note that similar conclusions were drawn about the ontogenesis of negative affective expressions in the human infant. Nonhuman primates also display the same pattern. These are interesting but limited observations. Critical questions remain unaddresssed. How is such a transition from reflexive, involuntary display to volitional control accomplished? What factors mediate the transition from rather stereotypic patterns of response to the rich variety and subtlety of affective communication seen in maturity? There are probably several important factors. In nonhuman primates, maturation, increasing cortical control, experience with auditory and kinesthetic feedback, and observational learning all probably play an influential role. In humans it is likely that the same factors obtain. However, both the ability to modulate affective display and the demand for control over affective expression are probably greater in humans than in lower primates (Edman, 1973; Izard, 1977). The greater press for the control of emotional expression would suggest that there may be additional mechanisms mediating this process. There is evidence that contingent responding by social agents plays a critical role in the modulation of both vocal and facial affective displays. This literature is reviewed in the following section.

THE SOCIALIZATION OF AFFECT The finding of Lenneberg et al. (1965) that children of deaf parents showed a wider range of cry behavior than those with normally hearing

18

Malatesta

parents suggests that contingent responding by the caregiver can modulate infant affective expressions. Further evidence from more recent studies is now available. It appears that mothers actively select, respond to, and modify infant affects and that this occurs in the context of maternal affect "coaching" or "display instruction." The current literature on dyadic interaction, at least, seems to bear this out. One of the most salient features of mother--infant interaction is the sheer volume of affective play in which both engage (Stern, 1974; Trevarthen, 1979; Kaye, 1979). The mother is clearly a potent affective stimulus for the infant. Brazelton et al. (1974), in their observational study of mothers and their 1- to 4-month-old infants, report being able to distinguish whether an infant is attending to the mother or watching an object by looking at any part of the infant's body; all three affect channels--face, body, voice--are engaged in active affective responding: the arms and legs may pedal slowly, the face alerts, and there are fleeting smiles, grimaces, and vocalizations. With the elicitation of this wealth of affectvie behavior comes the opportunity for affective shaping. Maternal shaping appears to be exerted in a subtle way. Mothers actively engaged in face-to-face play with their infants frequently imitate their vocal and facial expressions. But there is something unique about maternal imitation. Maternal imitations are not perfect miming of infant expressions but have a shaping and leading edge. Kaye (1979), drawing from videotapes of mothers and infants in face-to-face play at 6, 13, and 26 weeks, distinguishes three types of modifying imitation: (1) Minimizing imitation: In this type the mother tries to interfere with the infant's expression; for example, if the infant begins to cry, the mother may go into a similar cry face but then immediately flash to a bright face; (2) Modulating imitation: Here the mother stays with the expression but tempers it in some qualitative way, such as softening a harsh whine; (3) Maximizing imitation: In this case the mother may exaggerate her infant's expression. We don't yet know the proportion of different types of imitation at any one developmental stage, much less their distribution over time and their distribution per affect type, although this would be valuable information to have. Maternal exaggeration, however, has been noted by other researchers (Brazelton et al., 1974; Stern, 1974; Tronick, Als, & Adamson, 1979). The configuration of the mother's face during such expressions is cartoonlike in its schematic exaggeration, and the pace of the expression is like slow-motion film. Such behavior would seem quite bizarre in the context of an adult-adult interaction but is perfectly suited for the infant's learning schemas of human expressiveness if the infant can be assumed to have a slower rate of information processing. The slow pacing is almost demanded; the fleetingness, intergradation, and subtlety that are

Infant Emotion

19

characteristic of adult-adult communication would place an unduly difficult decoding burden on the cognitively immature infant. Mothers also employ a different speech register around infants, i.e., a simplified, exaggerated pattern of speech known as baby talk or Ammensprache (Trevarthen, 1979; Helfrich, 1979). The mother's vocal interactions involve using an expanded range of pitch, and the sound intensity is richly and dramatically varied (Stern, 1974). Trevarthen suggests that baby talk obeys tacit rules of expression; however, these rules have yet to be specified in a systematic way. We might speculate that exaggerated intonational patterns help develop and "set" schemas for vocal affect just as exaggerated facial expressions may help develop and set facial affect schemas. Another means of sharpening awareness is through the use of discriminative verbal labels for different facial expressions and moods; interestingly, mothers of 2-month-olds actively interpret the infant's affect state in the ongoing patter of the verbal banter of baby talk (Trevarthen, 1979). Schema setting for both facial and vocal affect may in fact be one means of developing the young infant's affect awareness or affect consciousness. Since presumably some level of reflective self-consciousness is a prerequisite to modulating raw affect, the mother's behavior plays an important part in helping the infant develop a greater degree of voluntary, instrumental control of his or her own affective responsiveness. What this suggests, then, is that affect modulation is importantly structured by maternal affect coaching, a process that begins very early in life. In summary, mothers engage in two types of behavior that would seem to have potential for influencing infant emotional development in important ways: modifying imitation and verbal interpretation. What we don't know is how the pattern of affect coaching may vary as a function of certain maternal and infant variables such as age and sex of infant and maternal emotional makeup, and whether some mothers selectively respond to one mode of emotional communication (for example, the vocal) over another. Clearly there is much important work to be done in this area before the process of infant affective socialization is more clearly understood.

CONCLUSION Phyletic as well as ontogenetic considerations suggest that there is early functional integrity among the vocal, facial, and bodily expressions of the most basic (i.e., survival-related) affects. Early infant affective expression appears to be a "whole-body" phenomenon, especially in the expression of negative feelings. Dissociation among the expressive channels

20

Malatesta

becomes increasingly more possible and likely, owing to neuromuscular maturation and, later, increasing voluntary (cortical) control. A comparison of the ontogenetic course of vocal affect expression in human and infrahuman primate young indicted many similarities. Displeasure sounds are chronologically and functionally more primary in development than the pleasure sounds and, at least early in development, appear to encode more variation in affective meaning. Both innate and learned factors play a part in the ontogenesis of the vocal expression of affect in both human and nonhuman primates. Data on the emergent properties of affective vocalization would seem to indicate that there may be certain universal vocal signals, as is the case with facial expressions. Vocal as well as facial display behavior, however, is subject to modification through certain maturational and learning experiences. In the case of the human infant, social factors are thought to play an important part in the modification of display parameters. As the infant matures he becomes increasingly engaged in dyadic interaction with the caregiver, and the raw affect of early infancy is tempered. Feedback and coaching, in the form of modifying imitation and verbal interpretation by the mother, are seen as instrumental in the infant's growing capacity for affect consciousness, moderation, and tolerance, although these processes are imperfectly understood. However, an explication of these processes may be one of the more important goals of developmental psychology. Affect shaping begins at a time when the central nervous system is most plastic and malleable. With maturation of higher cortical centers, the rules of affect expression as interpreted by the infant's caregiver superimpose control on the more primitive substrate of the phylogenetically older "emotional" brain; these first rules may be some of the most basic and firmly imprinted rules the child ever learns.

REFERENCES Ahrens, R. Beitrag zur Entwicklung des Physiognomie und Mimikerkennens. Zeitschrift far Experimentelle und Angewandte Psychologie, 1954, 2, 412-454. Aldrich, C. A., Sung, J., & Knopp, C. The crying of newly born babies. Journal of Pediatrics, 1945, 26, 313. Andrew, R. J. The origin and evolution of the calls and facial expressions of the primates. Behavior, 1963, 20, 1-107. Bastian, J. Primate signalling systems and human language. In I. Devore (Ed.), Primate behavior: Field studies o f monkeys and apes. New York: Holt, Rinehart & Winston, 1965. Brandt, E., Baysinger, C., & Mitchell, G. Separation from rearing environment in motherreared and isolation-reared rhesus monkeys (Macaca mulatta). International Journal of Psychobiology, 1962, 2, 193-204.

Infant Emotion

21

Brazelton, T. B., Koslowski, B., & Main, M. The origins of reciprocity: The early motherinfant interaction. In M. Lewis & L. Rosenblum (Eds.), The effect of the infant on its caregiver. New York: Wiley, 1974. Bridges, K. B. Emotional development in early infancy. Child Development, 1932, 3, 324-341. Bronson, G. W. The fear of novelty. Psychological Bulletin, 1968, 69, 350-358. Charlesworth, W. R., & Kreutzer, M. A. Facial expression of infants and children. In P. Ekman (Ed.), Darwin and facial expression. New York: Academic Press, 1973. Chevalier-Skolnikoff, S. Facial expression of emotion in nonhuman primates. In P. Ekman (Ed.), Darwin and facial expression. New York: Academic Press, 1973. Darwin, C. The expression o f emotions in man and in animals. Chicago: University of Chicago Press, 1965. (Originally published, 1872) Darwin, C. A biographical sketch of an infant. Mind, 1877, 2, 292-294. Dennis, W., & Dennis, M. G. Behavioral development in the first year as shown by forty biographies. Psychological Record, 1937, L 349-361. Ekman, P. Cross-cultural studies of facial expression. In P. Ekman (Ed.), Darwin and facial expression. New York: Academic Press, 1973. Ekman, P., & Friesen, W. V. Constants across cultures in the face and emotion. Journal o f Personality and Social Psychology, 1971, 17, 124-129. Ekman, P., & Friesen, W. V. Unmasking the face. Englewood Cliffs, New Jersey: PrenticeHall, 1975. Ekman, P., Sorenson, E. R., & Friesen, W. V. Pan-cultural elements in facial displays of emotion. Science, 1969, 164, 86-88. Emde, R. N., Gaensbauer, T. J., & Harmon, R. J. Emotional expression in infancy: A biobehavioral study. Psychological Issues, 1976, 10, 1-286. Fairbanks, G. An acoustical study of the pitch of infant hunger wails. Child Development, 1942, 13, 227-232. Green, S. Variation of vocal pattern with social situation in the Japanese monkey (Macaca fascata): A field study. In L. A. Rosenblum (Ed.), Primate behavior: Developments in field and laboratory research. New York: Academic Press, 1975. Helfrich, H. Age markers in speech. In K. R. Scherer & H. Giles rEds.), Social markers in speech. Cambridge: Cambridge University Press, 1979. Hinde, R. A., Rowell, T. E., & Spencer-Booth, Y. Behavior of socially living rhesus monkeys in their first six months. Proceedings of the Zoological Society o f London, 1964, 143, 609-649. Itani, J. Vocal communication of the wild Japanese monkeys. Primates, 1963, 4, 11-66. Izard, C. E. The face of emotion. New York: Appleton-Century-Crofts, 1971. Izard, C. E. Human emotions. New York: Plenum Press, 1977. Izard, C. E. On the ontogenesis of emotions and emotion-cognition relationships in infancy. In M. Lewis & L. A. Rosenblum (Eds.), The development of affect. New York: Plenum, 1978. Izard, C. E. & Buechler, S. Emotion expressions and personality integration in infancy. In C. E. Izard (Ed.), Emotions in personality and psychopathology. New York: Plenum press, 1978. Kaplan, E., & Kaplan, G. The prelinguistic child. In J. Eliot (Ed.), Human development and cognitive processes. New York: Holt, Rinehart & Winston, 1971. Kaye, K. Thickening thin data: The maternal role in developing communication and language. In M. Bullowa (Ed.), Before speech. Cambridge: Cambridge University Press, 1979. Lenneberg, E., Rebelsky, F., & Nichols, I. The vocalization of infants born to deaf and hearing parents. Human Development, 1965, 8, 23-37. Lewis, M. Issues in the development of fear. In I. L. Kutash & L. B. Schlesinger (Eds.), Pressure point: Perspectives on stress and anxiety. San Francisco: Jossey-Bass, 1979. Marler, P. Communication in monkeys and apes. In I. Devore ~ d . ) , Primate behavior: Field studies o f monkeys and apes. New York: Holt, Rinehart & Winston, 1965. Muller, E., Hollien, H., & Murry, T. Perceptual responses to infant crying: Identification of cry types. Journal o f Child Language, 1974, 1, 89-95.

22

Malatesta

Murray, A. D. Infant crying as an elicitor of parental behavior: An examiantion of two models. Psychological Bulletin, 1979, 86, 191-215, Murry, T., Amundson, P., & Hollien, H. Acoustical characteristics of infant cries: Fundamental frequency. Journal o f Child Language, 1978, 4, 321-328. Oster, H. Facial expression and affect development. In M. Lewis & L. A. Rosenhlum (Eds.), The development of affect. New York: Plenum, 1978. Plooij, F. How wild cimpanzee babies trigger the onset of mother-infant play--and what the mother makes of it. In M. BuUowa (Ed.), Before speech. Cambridge: Cambridge University Press, 1979. Prechtt, H. F. R., Theorell, K., Gramshergen, A., & Lind, J. A statistical analysis of cry patterns in normal and abnormal newborn infants. Developmental Medicine and Chitd Neurology, 1969, 11, 142-t52. Redican, W. K. Facial expression in nonhuman primates. In L. A. Rosenblum 0~d.), Primate behavior: Developments in field and laboratory research. New York: Academic Press, 1975. Ricks, D. Making sense of experience to make sensible sounds. In M. Bullowa (Ed.), Before speech. Cambridge: Cambridge University Press, 1979. Ringel, R., & Kluppel, D. Neonatal crying: A normative study. Folia Phoniatrica, 1964,

16, 1-6. Rowell, T. E. The social development of some rhesus monkeys. In B. M. Foss (Ed.), Determinants o f infant behavior. (Vol. 2). London: Methusen, 1963. Scherer, K. R. Nonlinguistic vocal indicators of emotion and psychopathology. In C. E. Izard (Ed.), Emotion in personality and psychopathology. New York: Plenum Press, 1979. Sherman, M. The differentiation of emotional response in infants. I. Judgments of emotionai response from motion picture views and from actual obsevation. II. The ability of observers to judge the emotional characteristics of the crying infant and of the voice of an adult. Journal o f Comparative Psychology, 1927, 7, 265-284; 335-351. Sherman, M. The differentiation of emotional responses in infants. III. A proposed theory of the development of emotional response in infants. Journal o f Comparative Psychology, 1928, 8, 385-394. Spitz, R. A. The first year o f life. New York: International Universities Press, 1965. Sroufe, L. A., & Waters, E. The ontogenesis of smiling and laughter: A perspective on the organization of development in infancy. Psychological Review, 1976, 83, 173-189. Sroufe, L. A., & Wunsch, J. P. The development of laughter in the first year of life. Child Development, 1972, 43, 1326-1344. Stark, R. E. Features of infant sounds: The emergence of cooing. Journal o f Child Language, 1978, 5, 379-390. Stark, R. E., rose, S. N., & McLagen, M. Features of infant sounds: The first eight weeks of life. Journal of Child Language, 1975, 2, 205-221. Steklis, H. D., & Raleigh, M. J. Behavioral and neurobiologiai aspects of primate vocalization and facial expression. In H. D. Steklis and M. J. Raleigh (Eds.), Neuroblology o f social communication in primaries. New York: Academic Press, 1979. Stern, D. The goal and structure of mother-infant play. Journal o f the American Academy o f Child Psychiatry, 1974, 13, 402-421. Takeda, R. Development of voad communication in man-raised Japanese monkeys I. From birth until the sixth week. Primates, 1965, 6, 337-380. Takeda, R. Development of vocal communication in man-raised Japanese monkeys II. From the 7th to the 30th week. Primates, 1966, 7, 73-116. Tomkins, S. Affect, imagery, consciousness: Vol. I: The positive affects. New York: Springer, 1962. Tomkins, S. Affect, imagery, consciousness: Vol II: The negative affects. New York: Springer, 1963. Tonkova-Yampol'skaya, R. V. Development of speech intonation in infants during the first two years of life. Soviet Psychology, 1968, 14, 94-101.

Infant Emotion

23

Trevarthen, C. Communication and cooperation in early infancy: A description of primary intersubjectivity. In M. Bullowa (Ed.), Before speech. Cambridge: Cambridge University Press, 1979. Tronick, E., Als, H., & Adamson, L. Structure of early face-to-face communicative interaction. In M. Bullowa (Ed.), Before speech. Cambridge: Cambridge University Press, 1979. Valanne, E., Vuorenkoski, V., Partanen, T. J., Lind, J., & Wasz-Hockert, O. The ability of human mothers to identify the hunger cry signals of their own newborn infants during the lying-in-period. Experientia, 1967, 23, 768. van Hoof f, J. A. R. A. M. The facial displays of the catarrhine monkeys and apes. In D. Morris (Ed.), Primate ethology. New York: Anchor Doubleday, 1969. Washburn, R. W. A sutdy of the smiling and Laughing of infants in the first years of life. Genetic Psychology Monographs, 1929, 6, 398-537. Wasz-Hockert, O., Lind, J., Vuorenkoski, V., Partanen, T., & Valanne, E. The infant cry: A spectrographic and auditory analysis. Clinics in Developmental Medicine, 1968 (29, Spastics International Medicine Publications), 1-41. Wasz-Hockert, O., Partanen, T., Vuorenkoski, V., Michelsson, K., & Valanne, E. The identification of some specific meanings in newborn and infant vocalization. Experientia, 1964, 20, 154. Wasz-Hockert, O., Partanen, T., Vuorenkoski, V., Valanne, E., Michelsson, K. Effect of training on ability to identify preverbal vocalizations. Developmental Medicine and and Child Neurology, 1964, 6, 393-396. Weeks, T. E. Speech registers in young children. Child Development, 1971, 42, 1119-1131. Wiesenfeld, A. R., Malatesta, C. Z., & DeLoach, L. L. Differential parental response to familiar and unfamiliar infant distress signals, lnfant Behavior and Development, 1981, 2, in press. Wilder, C. N., & Baken, R. J. Some developmental aspects of the infant cry. Journal of Genetic Psychology, 1978, 132, 225-230. Wolff, P. H. The natural history of crying and other vocalizations in early infancy. In B. M. Foss (Ed.), Determinants of infant behavior (Vol. 4). London: Methuen, 1965. Wolff, P. H. The causes, controls and organization of behavior in the neonate. Psychological Issues, 1966, 5, 1-105. Young, G., & Decarie, T. G. An ethology-based catalogue of facial-vocal behavior in infancy. Animal Behavior, 1977, 25, 95-107. Zeskind, P. S., & Lester, B. M. Acoustic features and auditory perceptions of the cries of newborns with prenatal and perinatal complications. Child Development, 1978, 49, 580-589.