Poster group
Details of individual items:
poster
The purpose of our cross-linguistic longitudinal (3, 6, 9, 12 months) study is to examine the joint influence of genetic and social factors in early language development for healthy Finnish and Russian children. The present poster presents vocalization data of four Finnish and four Russian infants. The infants' speech sounds were recorded in face-to-face and play interactions with the mother at 3 and 6 months of age. The temporal and the spectral characteristics of the acoustic sounds were estimated by CSL (ver. 4.0) and Cool edit 96 (Syntrillium Soft.Corp.). The instrumental analyses of infants vocalizations revealed both similarities and differences in spectral characteristics of 3- and 6-month-old infants' vowel-like sounds compared with adults' vowels. The results indicated that infants' vocal productions at 3 months of age contained sounds common for both Finnish and Russian languages. Comparative spectographic analyses did not reveal any significant differences in temporal and spectral characteristics of vowel-like sounds for infants exposed to different languages during their first months of life. It was found that the two first formants, which serve as 'acoustic cues' for differentiation of vowels, could not always be found in the infants' earliest vocalizations at 3 and 6 months. More higher values in fundamental frequency and, correspondingly first formant frequencies, when data obtained for adults, according literature, were revealed in vocalizations for both 3-and 6-month-olds infants. Our results suggest that, between 3 and 6 months of age, the speech environment has influenced vocal development in the direction of native language vowel categories. By 6 months of age, both Finnish and Russian infants' vocal productions appeared characteristic for their native language. In general, at this age the infants produced significantly more sounds and sound patterns in face-to-face interactions, than during play interactions or when distressed. We will discuss the implications of these findings with regard to those of other cross-linguistic studies.
poster
no abstract
poster
The capacity to relate two or more words into single intonational contoursis often referred as an important phenomenon characterizing the passagefrom single-word to early multi-word utterances (Bloom, 1973; Dore,46ranklin, Miller and Ramer 1976). Nevertheless, systematic investigationsof acoustic properties of early multi-word utterances are scarce. Earlyresearch on this topic (Branigan,1979; Fonagy,1972) shows that acousticproperties of successive single-word utterances evolves in the transitionto multiword utterances, but no information is offered about thepossibility that the semantic relationship between the two words is partlyresponsible for these changes. Final syllable lengthening has also beenstudied recently in a cross-linguistic perspective. According to Snow(1994) this phenomenon appears in English-speaking children afterdevelopment of the capacity to combine more words in a single sentence,while Konopczynsky (1991) observed it in French-speaking children in apre-syntactic phase.The aim of this work is to analyze the prosodic characteristics of earlymultiword utterances produced by a sample of 32 Italian children (15females). Children were recruited at thirteen/fourteen months andvocabulary development was assessed monthly up to c.200 words. At a fixedvocabulary size (50, 100 and 200 words) children and mothers participatedin a free play session.Children's productions were transcribed by a first coder with a precisionof 0.1 sec. Contextual information was also annotated. Reliability ofcoding was assessed by a second coder. We will only focus on the prosodiccharacteristics of the different types of multiword utterances, which weredivided into 1) Chained Successive Single Word Utterances (Bloom,1973)where each utterance refers to its own event or activity 2) Combinations inwhich the meaning of the whole utterance is different from the meaning ofsingle words. Intonational and timing characteristics of all multiwordutterances were analyzed using Winpitch software. We hypothesized thatdifferent prosodic models are detectable in the first combinationsdepending on semantic complexity and that final syllable lengthening onlyaccompanies production of more evolved multiword utterances.A preliminary analysis shows that the capacity to link two or more words insingle intonational contours is not present in all the first combinations.The coding of simple semantic relationships is realized from the beginningwith a single intonational contour, while more complex relationships arerealized through different prosodic models.ReferencesBloom, L. (1973). One word at a time: the use of single words utterancesbefore syntax. The Hague: Mouton.Branigan, G. (1979) Some reasons why successive single word utterances arenot. Journal of Child Language, 6, 411-421.Dore, J., Franklin, M. B., Miller, R. T., Ramer, A. L. H. (1976).Transitional phenomena in early language acquisition. Journal of ChildLanguage, 3, 13-29.46onagy, I. (1972). A propos de la genese de la phrase enfantine. Lingua,30, 31-71.Konopczynski, G. (1991). Le language Emergent II: Aspects vocaux atmE9lodiques. Hamburg: Helmut Buske Verlag.Snow, D. (1994). Phrase-final syllable lengthening and intonation in earlychild speech. Journal of Speech and Hearing Research, 37, 831-840.
poster
no abstract
poster
Recent studies suggest that the learning mechanisms involved inlexical development form the basis for children's acquisition of grammar(e.g., Bates; et al., 1988; Bates & Goodman, 1997, Akhtar & Tomasello,1999). Using the MacArthur Communicative Development Inventories (CDI),Fenson, et al. (1994) demonstrate that reported vocabulary productionbears a strong, but non-linear relationship to use of grammatical forms(r .84) in a large sample of children 16-30 months of age. Bates &Goodman (1997) report similar non-linear relationships in a longitudinalsample followed over the same developmental period. While these findingsare suggestive of common underlying mechanisms of learning, suchpatterns could also be attributable to general individual differencesacross children (i.e., children who are good at lexical learning aregood at learning grammar). To explore for this possibility, we examinedlexical and grammatical relationships in a sample of balanced'bilinguals,' i.e., children who are simultaneously learning Spanish andEnglish under the age of three years. Participants were healthy, full-term toddlers involved in an ongoingstudy of language development in Spanish- and English-speaking children.Children were excluded if they were reported to have significant medicalproblems, chronic otitis media, hearing loss, or developmentaldisability. Children (n 56) included in the current analyses (M 23.5months, SD 4.2) had 'balanced' input (M 48.5% English, 51.5%Spanish), based on a detailed phone interview. Individual(s) who werefamiliar with the child's language use completed the CDI: Words &Sentences and Inventario del Desarrollo de Habilidades Comunicativas:Palabras y Enunciados (Fenson, et al., 1994; Jackson-Maldonado, et al.,1993). On both forms, vocabulary production is indexed by number ofwords reported on the vocabulary checklist. Grammatical development wasmeasured by: (1) Score on a 37-item Sentence Complexity scale ('whatsounds more like what your child says,' e.g., 'kitty sleep' vs. 'kittyis sleeping'), and (2) Mean of Three Longest Sentences (M3L) that 'youhave heard your child say recently.' Although vocabulary production fell significantly below expectedlevels when English (M 31.2 percentile, t(51) 4.6, p < .001) andSpanish (M 25.9 percentile, t(51) 6.8, p < .001) were analyzedseparately, overall vocabulary growth was within the normal range whenboth languages were taken into account, M 47.1 percentile, t(51) .71, p .48 (Pearson, et al., 1993). Several significantcross-language, within-domain correlations were found: English x SpanishVocabulary: r .42, p < .002; English x Spanish Complexity: r .43, p< .01; English x Spanish M3L: r .33, p < .001. Interestingly,within-language, cross-domain correlations were similar in magnitude tothose observed in monolingual children (Vocabulary x Complexity inEnglish: r .87, p < .0001; Spanish: r .81, p < .0001). These dataindicate that there are striking similarities in the patterns of lexicaland grammatical acquisition of monolingual and bilingual children.Further, there is evidence that grammatical development is specificallytied to vocabulary growth in a given language, rather than to languagelearning ability more generally.
poster
no abstract
poster
Otitis media (OM) is one of the most common diseases of early childhood and frequently results in a mild-to-moderate hearing loss. These factors place children at risk for not obtaining the auditory-verbal experience necessary for normal phonetic development. In comparison to their healthy peers, differences have been found in the number of consonants used in babbling and the number of vocalizations containing consonants in comparison to vowels (Luloff et al., 1991; Robb et al., 1993). These studies point to the need for continued and more detailed studies of the relationship between OM and early speech development. The objective of the present study was to examine the influence of chronic otitis media (COM) on the syllable structure and the range and types of consonants produced in the babbling of children with COM at age 12 months. Speech samples of 40 children, 20 with COM and 20 with negative histories of OM were examined at age 12 months. Children wore a very small lavalier microphone connected to a wireless FM transmitter linked to a FM receiver. Samples were videotaped during a 15-minute play period on a mobile van equipped with a playroom. All children were Caucasian and entered daycare during their first year of life. Hearing testing was conducted in the mobile van using visual reinforcement audiometry. Middle ear functioning was assessed using pneumatic otoscopy and immittance audiometry and children were assigned a weekly diagnosis. Twenty children with OM a minimum of 30% of the time during 12 months were classified as COM and 20 were classified as low OM (ill less than 30% of the year).Vocalizations were glossed and transcribed phonetically according to the conventions of the International Phonetic Alphabet. Results were described in terms of the quantity and diversity of consonants produced and the complexity of syllable structures used. The number of consonant types, the proportion of vocalizations containing a true consonant (Vihman & Greenlee, 1987), and mean babbling levels (Stoel-Gammon, 1992) were calculated. Results showed the babbling repertoires of children with chronic otitis media were significantly more restricted than those of their healthy peers in measures of syllable structure and consonant types. Children in the nonchronic group used more true consonants than the chronic group. Babbles of the healthier group contained variegated and canonical babbles. Phonetic inventories were similar to those reported for normally developing 12 month olds (Robb & Bleile, 1994). Children with COM, however, had atypical inventories that contained simpler syllable structures, fewer true consonants and an abundance of dorsal and glottal articulations. Results of audiologic testing, pneumatic otoscopy, and tympanometry indicated children with COM had the most elevated hearing thresholds. Children with the earliest episodes of OM were the most affected. Results suggest a relationship among COM, hearing and early phonetic development. Descriptions of young children's babbling inventories may provide valuable diagnostic measures for the early identification of children with COM who are at risk for delays in other aspects of language development.
poster
Language and cognition literature has shown that early communicativegestures, shared attention and social routines are related to subsequentlexical achievements (Iverson & Thal, 1998; Bates et.al., 1975; Rollinset.al., 1998; Capirci et.al., 1996). Particularly, deictic gestures play animportant role in later language development. Deictic words are usuallyearly and include forms like 'this', in English or 'qua' in Italian. Wordsthat call attention to an object or event, such as 'guarda' and 'look' mayappear early, but are not commonly used. In Spanish, though, the word'mira' is a very early and frequent form. Further analysis of 'mira' mayaid in the understanding of the function of shared attention and deixis byadding a verbal form of communicative signaling. The purpose of this studywas to determine whether the word 'mira' serves a verbal deictic functionsimilar to other communicative gestures. This paper presents the relationship between early communicative andsymbolic gestures to lexical development in thirty 12 month old Spanishspeaking children living in Mexico. Data was obtained from natural languagesamples of children interacting with the parent and/or researcher forapproximately 30 minutes. Total number of utterances and gesture types(communicative, routine, symbolic) were registered. Occurrences of 'mira'