Poster workshop
Chair: Graham Schafer
Discussant: Margaret Harris
The five posters presented here all address the issue of the extent to which early language comprehension may be governed byÑrather than simply Ôtaking account of'Ñearly experience. The contributors use a similar paradigm (preferential looking: Golinkoff, Hirsh-Pasek, Cauley, & Gordon, 1987), in a number of different laboratories, to test infants' sensitivity to various aspects of the environment, manipulated a priori by the experimenter.We explore the early language development of infants (9 to 28 months) in terms of specific measures of their experience in different domains. For instance, participants' experience of (and hence knowledge about) objects, speech, gaze, joint naming routines, and other statistical regularities in the environment are manipulated by the experimenters. Infants' receptive responses to language are then measured in a closely controlled laboratory environment, where the effects of experienceÑboth recent, long term, and cross-sequentiallyÑcan be assessed. Long term factors include measures of vocabulary development, and also longitudinal training intervention with specific words and referents. Short-term interventions take the form of periods of training in the laboratory immediately before testing, in which infants are presented with pairs of images where one is consistently named and one is not, or where one image is named by a Ôface' and one is named without gaze, or where infants are presented with word-object associations relying on minimal or non-minimal contrasts. In testing, we establish the contribution of these short- and long-term manipulations, as well as examining the effects of departures from prototypical representations. The contribution of this group of posters is thus richly interconnected, and opportunities exist for exchanges on theoretical, empirical, and methodological terms.Potential discussion points: (1) Three of these posters concentrate on the issue of consistency (covariation), in terms of either training, or departures from prototypicality. (2) The posters all speak, in different ways, to the currently important issue of the influence of naming on categorization and vice versa. (3) To what extent does vocabulary size mediate early-emerging lexical behaviors? Such ideas may form the basis of a discussion of the potential explicatory value of such mechanism(s) for the understanding of the phenomena of early vocabulary development.Golinkoff, R.M., Hirsh-Pasek, K., Cauley, K.M., & Gordon, L. (1987). The eyes have it: Lexical and syntactic comprehension in a new paradigm. Journal of Child Language, 14, 23-45.
Details of individual items:
poster
Which referents are children willing to accept for words they know? How does the scope of a word change over time? To answer these questions, early word comprehension was assessed for object and animal words as well as for prepositions.Prior to all three studies children's experience with the words in question was investigated by using a British version of the Communicative Development Index (CDI). Using the Preferential Looking task children were then shown two stimuli side-by-side, a target and a distracter, and heard the target stimulus named. Target stimuli were either typical or atypical exemplars of the named category. We predicted that children first connect typical examples with the target name, and broaden the extension of the name as they get older to include less typical examples.For object and animal words, the first study shows that when targets are named, 12-month-olds display an increase in target looking for typical but not atypical targets whereas 24-month-olds display an increase for both.The second study shows that 18-month-olds display a similar pattern to 24-month-olds. These results confirm that 12-month-olds do indeed restrict the scope of their object and animal names to typical exemplars while children from 18 months onwards have broadened their scope to include less typical examples.In a third study we investigated which spatial prepositions are the first to be comprehended by 18- and 24-month-olds and how they are understood by different age groups. Here, children were shown stimuli of objects or animals on or under a table in typical Ôon'- and Ôunder' situations (cup in middle of table / cup centrally under a table) versus atypical Ôon'- and Ôunder' situations (cup on table at table edge / cup under table at table edge). We predicted that children do indeed comprehend the prepositions Ôon' and Ôunder' very early as indicated by existing research and parental assessments. Furthermore, we predicted that children's looking behaviour shows typicality effects similar to those in the object word studies above. The results corroborate that Ôon' and Ôunder' are among those prepositions that are understood early and also that children do indeed differentiate between typical and atypical Ôon' and Ôunder'-situations in their early comprehension of these prepositions. Further implications for the early development of word comprehension will be discussed.
poster
Orienting to objects by infants has been shown to be mediated by their recent pragmatic-linguistic experience. Baldwin and Markman (1989) showed how labeling can direct infants' attention towards a novel item. Twelve-month-olds looked more at an object as it was being labeled (Experiment 1), and both 12- and 18-month-olds looked more at an object after it had been pointed at and labeled in comparison with a condition in which it had merely been pointed at (Experiment 2). Baldwin and Markman noted several candidate loci for attentional facilitation by labeling, one of which was sensitivity to covariation (p. 396). Subsequently, Baldwin has written trenchantly against the notion of the infant as a covariation detector, at least in the domain of language (Baldwin, 1995). Here, data are presented suggesting that infants of 15 months may use covariation information in order to orient towards objects.Fifteen-month-old infants were presented with novel images and novel words, in a laboratory environment. The experiment was divided into two phases. In the training phase, infants viewed one of two images in isolation, with an accompanying label. In the case of one of the images, the same label was repeated each time the image was seen (twelve times during the experiment). In the case of the other image, a different label was used each time the image was heard. In the testing phase, infants viewed both images simultaneously. There were two auditory conditions which were presented within-subjects. In one condition, infants viewed the image pair in silence. In the other , infants heard 'Look! Look!'. Infants looked more at the consistently-named object in the 'Look!' condition than in the silence condition.The behavior of the infants is discussed with respect to various theoretical accounts of infants' orienting to recently encountered images, including networks implementing the Rescorla-Wagner theory of classical conditioning (Rescorla & Wagner, 1972). It is concluded that the data are consistent with a parsimonious approach to language acquisition in which the infant uses statistical regularity in orienting to, and learning about, words, though this is not the only construal of the data.Baldwin, D. A. (1995). Understanding the link between joint attention and language. In C. Moore & P.J. Dunham (Eds.), Joint attention: Its origins and role in development.Baldwin, D. A., & Markman, E. M. (1989). Establishing word-object relations: A first step. Child Development, 60, 381-398.Rescorla, R.A., & Wagner, A.R. (1972). A theory of Pavlovian conditioning: Variations in the effectiveness of reinforcement and non-reinforcement. In A.H. Black & W.F. Prokasy (Eds.) Classical conditioning II. New York: Appleton-Century-Crofts.
poster
A six-month longitudinal study of 9 Ð 15 month-old infants is being undertaken to investigate the effects of parental intervention on language comprehension. A total of 62 infants (average age 39 weeks at commencement of the study) have been allocated to one of two experimental groups, or to a control group. All groups are matched for age, sex and family circumstances. Parents in all groups are asked to complete an Infant Language Questionnaire based on the MacArthur Communication Development Inventory at 9, 12 and 15 months.The experiment is divided into two phases. Each phase consists of a three-month training period, followed by a laboratory test. For the training period, parents in the two experimental groups are provided with specially designed stimuli and detailed instructions on how to conduct a structured learning programme with their baby in a domestic environment. The training stimuli consist of two sets of picture books and laminated picture cards (one set per experimental group). Each set of training stimuli illustrates eight different words, represented by six exemplars of the corresponding image. Stimuli and training methods have been chosen to reflect known or conjectured principles employed by infants during early word learning (see e.g., Woodward & Markman, 1997).Testing will be conducted by means of a preferential looking task (Golinkoff, Hirsh-Pasek, Cauley & Gordon, 1987). During the testing session, infants will be presented with a pair of images simultaneously on adjacent monitor screens, and auditory stimuli over a loudspeaker. The visual stimulus pairings will be composed of a novel exemplar of an image from each of the experimental groups (i.e. one trained, one untrained). Two sets of trials are planned:1. A non-referential task where the infant is instructed simply to look.2. A referential task where the name of a target image is embedded in continuous speech.Potential experimental effects controlled for include left/right looking preferences and the effects of training during the testing session. The duration of infants' looking times to the left and right will be blind scored off-line by analysing videotapes of each testing session.The study will test the following hypotheses:· Orienting to a novel image is mediated by regular exposure to images drawn from the same basic-level category· Parental intervention can improve lexical comprehension performance· Training can sensitise infants to specific wordsReferencesGolinkoff, R.M., Hirsh-Pasek, K., Cauley, K.M., Gordon, L. (1987). The eyes have it: Lexical and syntactic comprehension in a new paradigm. Journal of Child Language, 14, 23-45Woodward, A.L. & Markman E.M. (1997). Early word learning. In D. Kuhn & R.S. Siegler (Eds), Handbook of Child Psychology, Volume 2: Cognition, perception and language. New York: Wiley
poster
Recent research has demonstrated that infants at 14, but not 8 or 20months, have difficulty discriminating phonetically similar novel words,when one of the words has been associated with an image in a bimodalhabituation paradigm (Stager & Werker, 1997; Werker, Cohen, Lloyd,Casasola & Stager, 1998; Werker, personal communication). These findingshave been interpreted as evidence that infants, who are beginning to learnword-referent relationships, i.e. beginning to learn words, undergo aperiod of functional reorganisation, where limited cognitive resources areused to map words to their referents at the expense of resolving difficultcontrast discriminations (Stager & Werker, 1997). This study investigateswhether healthy 18-month-olds can learn phonetically dissimilar andsimilar novel words in a preferential looking paradigm. This technique isecologically closer to the word learning process than the bimodalhabituation paradigm because infants have to do more than detect a changein the word or image in the word-image pairing: they have to show a visualpreference for the referent as opposed to the distractor image, for thegiven word to demonstrate word learning. Preliminary results suggest that 18-month-olds can map a pair ofauditorily-presented words to their respective referents (images) whenwords are phonetically dissimilar, thus learning more than one bimodalword-referent representation at the same time. It is not unexpected thatthese infants appear to learn the phonetically dissimilar words moreeasily than the similar ones, given that the dissimilar words arephonetically more distinctiveness than the similar words. These findingsdo not refute the notion that 18-month-olds, like the 14-month-olds, mayalso experience functionally reorganisation. Further investigation has yetto be completed to determine whether these 18 month-olds learn words moreeasily when they have larger rather than smaller vocabularies, such asmore than 50 words in production. This may shed light on whether increasedlexicon size, suggested to precipitate a shift from a holistic to asegmental processing strategy (Walley, 1993), may be as important as age,in explaining why phonetically similar words seem to be learnt more easilyas children get older (e.g. Charles-Luce & Luce, 1995; Barton, 1980). Barton, D. (1980). Phonemic perception in children. In G. H.Yeni-Komshian, J. F. Kavanagh & C. A. Ferguson. Child Phonology:Perception (vol. 2). London, Academic Press. pp. 97-116.Charles-Luce, J. & Luce, P. A. (1995). An examination of similarityneighbourhoods in young children's receptive vocabulary. Journal of ChildLanguage, 22, 727-735.Stager, C. L. and Werker, J. F. (1997). Infants listen for more phoneticdetail in speech perception than in word-learning tasks. Nature,388(6640), 381-382.Walley, A. C. (1993) The role of vocabulary development in childrensspoken word recognition and segmentation ability. Developmental Review, 13(3), 286-350.Werker, J. F., Cohen, L. B., Lloyd, V. L., Casasola, M. and Stager, C. L.(1998). Acquisition of word-object associations by 14-month-old infants.Developmental Psychology, 34(6), 1289-1309.
poster
Several recent laboratory studies of lexical development in infancy have reported rapid word learning in children as young as 13 months (e.g. Woodward, Markman & Fitzsimmons, 1994; Schafer & Plunkett, 1998). However, studies finding such associations between novel objects and novel names after only minimal pairings of the two have not always been easy to replicate. A series of preferential looking experiments investigate several factors that might explain infants' difficulties in learning words in the laboratory. First, it was hypothesised that infants' lack of familiarity with the novel objects and words typically used in studies of word learning might hinder the learning process. In an initial study, therefore, objects that were familiar to infants, but whose names were not yet known to them, were presented to infants three times in association with their real labels in an attempt to show that Ôrapid word learning' can be even more rapid when infants are already familiar with the object being labelled. Second, the relative importance of the social context surrounding word learning was examined, specifically in terms of infants' ability to benefit from the joint attention of an adult. Children's ability to learn the association between an object and its label using only statistical regularities in the input was compared to learning when information from an adult's direction of gaze was also available. Infants were presented with pairs of images while hearing the name of one of them. In one condition, an animated image of a face looked at the target image that was to be associated with the auditory label, guiding children's inferences as to which object was the referent of a new word. In a second condition, no face was present, and infants could only infer the target of the label from statistical information about the co-occurrences between image and label. It was hypothesised that while word learning would be more robust when joint attention was available, word learning was nonetheless expected to occur when the only information available to the infant regarding the referent of a word was its relative frequency of occurrence with the target object.ReferencesSchafer, G. and Plunkett, K. (1998). Rapid word learning by 15-month-olds under tightly controlled conditions. Child Development, 69(2), 309-320.Woodward, A.L., Markman, E.M. and Fitzsimmons, C.M. (1994). Rapid word learning in 13- and 18-month-olds. Developmental Psychology, 30, 553-566.