Symposium
Chair: Patricia Zukow-Goldring
Discussant: Margaret Harris
This symposium focuses on how infants come to know 'what everyone else already knows'. Prior to comprehending and producing words, infants gradually notice, engage in, and communicate within an expanding range of settings and events. This panel takes seriously recent calls to investigate the interpersonal and physical environment in which the infant develops in order to illuminate these issues (Bloom, 1997; E. Gibson, 1997). Research based on the ecological approach to perceiving, acting, and knowing has generated an extensive body of empirical studies on infant perception and action (Dent-Read & Zukow-Goldring, 1997; Goldfield, 1995; Reed, 1996). In recent work building on J. J. Gibson's ecological realist theory (1979), perceiving is broadly conceived as knowing the world, while action is broadly conceived as including acts of both nonverbal and verbal communication.The participants in this symposium have examined how perceptual processes in common daily activities inform infants' ensuing communicative development. Although they may conduct their research using a variety of methods, all are examining the processes that guide knowing, especially the selecting and detecting of perceptual information. The panelists have investigated a range of topics that document ontogenetic changes in infants' responses to perceptual information. They have examined the methods caregivers use to make perceptual information or structure prominent and available to their infants, the nature of infant-caregiver interactions with respect to affordances or opportunities for action with objects, the role of amodal invariants in detecting the correspondence between gesture and speech, and the relation between conventional object use and symbolic development. These studies conducted in Spain and Switzerland as well as in the United States among Euro-American, Latino, and Asian-Indian families include observations in natural settings as well as experimental data from the laboratory. The ecological approach leads to a new view of what develops as well as how development takes place. Just as perceiving does not take place in the organism, development does not take place in the organism. Rather, development takes place in the relation of the organism to the environment through an open-ended, continual process of coordination. Our goal is to integrate the unique contribution of this work into a more complete and coherent ecological theory of development than now exists in the field of developmental psychology.
Details of individual items:
paper
Psychological theorists who address the prelinguistic period rarely conceptualize objects as entailing an analysis in terms of public meaning. Objects -referred to usually as the 'physical world'- are isolated from the social world, from conventions, and from communication with other people. Generally, only very formal, 'syntactical' characteristics are considered. As a consequence, very often symbols are placed inside the head and are considered essentially as representational tools. Semiotic theories are an important resource when attempting to account for these developmental issues. Peirce and Eco inform our approach. Peirce's theory of signs focuses on how meaning relates to the world. Umberto Eco pinpointed a limitation in other views recently by complaining that analytical philosophy is satisfied with the concept of truth itself, yet it has not addressed the issue of our prelinguistic relation with things (Eco, 1997). We will consider, from a semiotic perspective, how infants build meanings related to objects. These meanings will change as the infant develops. Accordingly, the use of objects (nonconventional, conventional and symbolic) will evolve as well. All these transformations occur before language appears as a system. To understand this aspect of ontogenesis, many questions need to be considered and answered. Some of them are: Where does the capacity to differentiate signifiers from signifieds in early symbols come from? What are the functions of those first symbols? What are their precursors? What is the relation between early symbols and other semiotic systems that are functional in earlier periods of development and those that start to appear in the same period, e.g., language? What is their relation to objects in the world? What is the link to the referent? Do those relations change throughout development? How does communication with other people relate to the appearance of first symbols? In previous work we examined the importance of placing objects in a pragmatic position, i. e., considering objects from the point of view of their use in order to understand development. We suggested that there is an important link between conventional uses and symbolic uses of objects. Even though each type of object use has a different function and a different configuration, they share important elements, nevertheless. If they did not, it would be difficult to grasp how anyone could understand children when they 'pretend' to do something.These questions will be discussed based on empirical data from an observational study involving Swiss and Spanish children from 9 months through 14 months of age in triadic interactions with an adult and an object.
paper
The phenomenon of joint visual attention, i.e., the intentional coordination of visual attention by two or more participants, represents one form of early infant communication. The importance of being able to look where another person is also looking is that these early communicative circumstances promote the sharing of information, as well as the infants' ability to learn the affordances of different objects or events. While some developmental accounts of joint visual attention emphasize infants' representational abilities, or the infant's ability to learn the intentional significance of an adult's finger points, change in direction of gaze, etc., we will offer an account of joint visual attention and early communication that emphasizes infants' perceptual abilities. The first factor we will address is the affective expression of the adult and its effects on 7-month-olds' participation in joint visual attention. When an adult poses a positive, i.e., 'happy' expression, or a negative, i.e., 'sad' expression infants' frequency of joint visual attention decreases. However, when the adult's affective expression is neutral, or without any affective expression, 7-month-olds establish significantly more instances of joint visual attention. Second, we will describe evidence documenting that 9-, 12-, and 18-month-olds' participation in joint visual attention increases as the gesture becomes more elaborate or redundantly specifies where the adult is looking. Specifically, infants at all ages participate in more joint visual attention when an adult looks and points toward the target than when the adult only looks toward the target. A third factor we will describe that is hypothesized to affect infants' participation in joint visual attention is the type of target. Our results indicate that infants engage in more instances of joint visual attention when the targets differ with regard to their shape and color on each trial than when the targets do not differ across trials. Briefly, these results suggest that infants' participation in joint visual attention is affected by their ability to attend to the attention-directing gesture, detect the location specified by the attention-directing gesture and whether the attention-directing gesture refers to a novel or possibly interesting target. The final experiment to be described is a re-examination of the claim that 18-month-olds more frequently learn a novel label for an object when the adult joins the infant's focus of visual attention compared to those situations where the infant aligns their visual attention to match the adults focus of visual attention. Our results suggest that label learning occurs with equal frequency in either case. In other words, an 18-month-old's ability to learn a novel label is promoted by the adult and the infant actually engaging in the joint visual attention episode itself; whether the infant's gaze was followed into, or redirected by the adult does not seem to be significant.
paper
How do infants discover the relation between a segment of continuous speech (a word) and a specific aspect of the dynamic perceptual structure (e.g., an object)? We propose a perceptually based theory, informed by ecological realism, to answer this question. This approach postulates that detecting amodal regularities in gesture and speech promotes the emergence of the lexicon. Early in life infants can detect the pairing of stimuli across two modalities when the temporal characteristics of the stimuli are synchronous (Bahrick, 1983; E. Gibson, 1979; Lewkowitz, 1994; Spelke, 1979). During the later part of the first year of life and during the one-word period, the infant's lexical development is linked to attention (Adamson & Bakeman, 1984; Tomasello, 1988). Our work examines whether the infant's early ability to detect amodal regularities explains the observed relationship between attention-directing during early communications and later lexical development. Zukow-Goldring (1996, 1997, in press) has suggested that the temporal synchrony of gestures and speech in caregiver messages underlies the attention-directing aspect of these messages and cultivates the emergence of the lexicon. Her theoretical approach is based on a Gibsonian analysis of perceptual information and perceptual development (J. Gibson, 1966, 1979; E. Gibson & Rader, 1979) and supported by longitudinal data collected monthly over a two-year period from Latino and Euro-American families.We examined experimentally the connection between lexical development and the synchronous embedding of gestures and speech within caregiver messages. The design allows a comparison between messages with dynamic gestures and ones with nondynamic gestures. Ten infants aged 10.5 - 14 months watched two videotaped scenarios of a female uttering a nonsense word as she presented a novel object. In one condition the actor waggled and loomed the object toward the observing infant in a dynamic 'show' gesture. In the other condition, she displayed a static, ' held' object accompanied by a different nonsense word. The two scenarios were otherwise identical. The combinations of object/gesture pairings and the order of presentations were counterbalanced across subjects. To control for any inadvertent effects of the caregiver's gaze, postural, or tactile cues, the infant sat in an infant seat rather than on the caregiver's lap. To determine comprehension, a testing phase followed in which a recorded voice requested that the infant look at first one and then the other object displayed to the left or right of the screen. The order of the requests and the positioning of the objects in the video were counterbalanced. The infants directed more of their looks at the correct object if it had been presented initially with a dynamic gesture.This study reveals that embedding gesture and speech in the same synchronous, dynamic message contributes more robustly to word comprehension than a static communication. We will discuss this finding in the context of a more comprehensive theory of language development and articulate its implications for therapeutic and educational practices.
paper
How do infants learn the conventional links between words and referents? I present four studies to show that preverbal infants learn and remember word-referent links by perceiving the temporal synchrony between adults' vocalizations and gestures. Study 1. In experiment 1 (1998: Fig. 1, Study 1) 7-month-olds learned the arbitrary links between two syllables, /a/ and /i/, and two distinct objects when temporal synchrony was provided between the vocalizations and the objects' motion during habituation. No learning occurred when the vowels were spoken out of synchrony with these moving objects or with static objects. The infants also remembered the pairings after 10 min (submitted), on a two-choice intermodal preference test if temporal synchrony was present during learning (t (13) 2.29, p < .05). In experiment 2, sixteen 7-month-olds learned these vowel-object pairs in the presence of temporal synchrony [t (15) 2.38, p < .05], and remembered them 4 days later [t (15) 2.57, p < .025]. Study 2. We tested infants' learning of the links between more complex syllables and identical objects. In the first experiment, 7- and 8-month-olds (16 of each age) were habituated to two alternating displays of the objects moving in temporal synchrony with /ta/ and /ga/. Following habituation, only the 8-month-olds indicated learning of the syllable-object pairs by lookinglonger to the switch trials than to the no-change trials (submitted). In a control experiment, when sixteen 8-month-olds were habituated to /ta/ and /ga/ out of synchrony with the moving objects, no learning occurred (Fig. 1, Study 2). These findings suggest that infants use temporal synchrony between spoken words and moving objects to discover word-world links. Study 3. Provided ecological support for this thesis (in press). Analyses of 5-min. play-episodes showed that, American (n 8), and Asian-Indian (n 6) mothers primarily named objects and actions in temporal synchrony with hand-held object motions for their 5- to 8-month-olds (Fig. 2, Study 1). Study 4. We asked twenty-four American mothers to teach their 6- to 7-month-olds the names, gow and chi, for two toys, a raccoon and a Martian, during a 3-min. play-episode. Again, mothers primarily named these words in synchrony with object-motions (Fig 2, Study 2). Next, we tested infants' learning of the word-object links using the two-choice intermodal preference procedure. The 12 infants who looked longest to these displays (median looking .89, range .28 - 1.0), looked first to the word-matched object on the first block of trials (M .63, SD .13, t (11) 3.39, p .006) and across both blocks (M .56, SD .096, t (11) 2.29, p .043). Mothers' temporal coordination likely facilitated these infants' lexical-mapping on test trials. The remaining 12 infants who attended less to the videos showed no learning of word-object links.Supporting the ecological view of lexical development, these findings suggest that preverbal infants discover word-referent links by perceiving caregivers' temporal coordination of vocalization and gesture. Complementarily, infants' language environment is designed to facilitate detection of word-world links (Gogate, Walker-Andrews, & Bahrick, in press; Zukow-Goldring, 1997).