This study examined mothers' accuracy in predicting the responses their children gave and the scores they achieved on two standardized vocabulary tests. Three groups of 16 mothers and their preschool children (specific language-impaired; age-matched, language-normal; and younger, language-matched, language-normal) completed the Peabody Picture Vocabulary Test-Revised, Expressive One-Word Picture Vocabulary Test-Revised, and Vineland Adaptive Behavior Scales. Mothers overestimated their children's standardized receptive and expressive scores, with the exception that the mothers' estimates of the receptive vocabulary scores for language-impaired children did not differ from the actual test scores. Mothers of age-matched normals were best able to predict the labels their children used to name various pictured items. However, the overall estimates by mothers of language-impaired children were more accurate than those by mothers of language-normal children.

Children in the United States and in other English-speaking countries often learn a good deal about letters before they begin formal reading instruction. We suggest that one important and previously unrecognized type of knowledge about letters is knowledge of the phonological structure of the letters' names. In two experiments, preschoolers with a mean age of 4;8 judged whether various syllables were letters. The children made significantly more false positive responses to syllables such as /fi/, which have a phonological structure shared by a number of letters, than to syllables such as /fa/ and /if/, which sound less like real letters. This was true even for children who could recite the alphabet without error. Learning the alphabet, we conclude, forms the basis for generalizations about the structure of letter names.

There has been a growing trend in recent years toward the attribution of adultlike syntactic categories to young, language-learning children. This has derived support from studies which claim to have found positive evidence for syntactic categories in the speech of young children (e.g., Valian, 1986). However, these claims contradict the findings of previous research which have suggested that the categories underlying children's early multiword speech are much more limited in scope (e.g., Braine, 1976). The present study represents an attempt to differentiate and test these models of early multiword speech: focusing on the syntactic category of determiner, we investigated the extent to which 11 children showed overlap in the contexts in which they used different determiner types in their early multiword corpora. The results demonstrated that, although children do use determiners with a semantically heterogeneous collection of different noun types, there is very little evidence that they know anything about the relationship between the different determiner types, and thus there is no real case for the attribution of a syntactic determiner category. Indeed, this pattern of determiner use seems perfectly consistent with a limited-scope formula account of children's early multiword speech, as proposed by Braine (1976). These findings suggest that the development of an adultlike determiner category may be a gradual process, one involving the progressive broadening of the range of lexically specific frames in which different determiners appear, and are broadly consistent with a number of recent constructivist models of children's early grammatical development.

The purpose of this study was to investigate early literacy acquisition in a phonologically transparent orthography with regular letter-sound correspondences. It was considered that Turkish, with its systematic phonological and orthographic structure, would make different demands on the beginning reader than the languages used in many of the previous studies of literacy acquisition. First grade children were assessed using tests of phonological awareness, letter recognition, word and pseudoword recognition, spelling, syntactic awareness, and listening comprehension at the beginning of the school year. The impact of these factors on the development of word recognition, spelling, and reading comprehension was examined. The results strongly suggest that a phonologically transparent orthography fosters the early development of word recognition skills, and that phonological awareness contributes to word recognition in the early stages of reading acquisition. Once the children's word recognition performance is high, listening comprehension ability distinguishes the different levels of reading comprehension among children. These patterns of results were interpreted as reflecting the phonological and orthographic characteristics of the Turkish language and orthography.

One of the most striking facts about exchange errors in speech is that open class items are exchanged, but closed class items are not. This article argues that a pattern analogous to that in speech errors also appears in intrasentential code-switching. Intrasentential code-switching is the alternating use of two languages in a sentence by bilinguals. Studies of the spontaneous conversation of bilinguals have supported the claim that open class items may be codeswitched, but closed class items may not. This claim was tested by two sentence repetition experiments, one with Japanese/English bilinguals and the other with Spanish/English bilinguals. The results show that the switching of closed class items caused significantly longer response times and more errors than the switching of open class items.

The word recognition processes of proficient bilinguals were examined in their mother tongue (Greek) and in English in relation to the linguistic and syntactic characteristics along which the two languages differ. Their processes were then compared with those of monolingual readers.The following issues were addressed: the nature of bilingual functioning, whether it is language specific, and the factors that affect second language reading development. These issues were examined within the context of a letter cancellation paradigm. The results indicated that bilingual readers performed differently in each of their two languages, conforming more the monolingual patterns in their mother tongue than to those in their second language. This discrepancy was interpreted as a lack of coordination of different word recognition skills in the second language.

In American Sign Language (ASL), in addition to manual signs, specific nonmanual behaviors play a crucial role in the grammar of the language. For example, conditionals and relative clauses are signaled by obligatory nonmanual markers. This study focuses on the acquisition of negation in ASL, which is signaled by manual signs as well as an obligatory headshake. In particular, we address the developmental relationship between the communicative and grammatical (or linguistic) headshakes for negation. Study 1 includes naturalistic data from a cross-sectional sample of 51 deaf children, ranging in age from 1: 0 to 4: 11, who are acquiring ASL as their primary language. Study 2 includes longitudinal data from 16 of these children.

The acquisition of the handshape aspect of American Sign Language signs was examined longitudinally in nine young children of deaf parents. In monthly home visit sessions, the parents demonstrated on videotape how their children formed the different signs in their lexicons. According to these parental reports, handshapes were produced accurately in 49.8% of the children's different signs. Accuracy of handshape production typically improved with the children's increasing age and vocabulary size. Four basic handshapes (/5, G, B, A/) predominated in the children's early sign productions. Measures of the children's handshape production accuracy, ordinal position of initial production, and frequency of production were used to describe the order in which handshapes were most often acquired. It was also observed that the part of the hand involved in contacting a sign's location often affected the accuracy of the handshapes being produced.

A new procedure called single-word shadowing was applied to the study of lexical access in context. Subjects listened to word pairs or sentences recorded in one voice and were asked to repeat the target word signaled by a voice shift. This technique yielded rapid and robust priming effects in normal adult subjects in word pairs and in a sentence context. Regression analyses showed that the semantic priming effects were large and significant, even when several additional factors believed to affect lexical access were controlled. Evidence was found for robust semantic priming in the healthy elderly and in children from 7 to 11 years of age, and there was also evidence for a change in the size and nature of context effects across the lifespan. Because single-word shadowing works across a broad age range and does not require reading, secondary tasks, or metalinguistic judgments, it is a promising tool for the study of lexical access in a range of different populations.

In this study, Spanish-English bilinguals and English monolinguals used brief English contexts to choose among possible meanings for unfamiliar words. Two types of errors were compared: transfer errors, which were answers consistent with Spanish, but not English, syntax, and nontransfer errors, which were inconsistent with the syntax of both languages. Nontransfer errors were found to be negatively correlated with reading proficiency in both Spanish and English. Transfer errors, on the other hand, were positively correlated with reading proficiency in Spanish and were unrelated to reading proficiency in English.

This study examined the conversational patterns of older children with their younger siblings. The sample included 26 nondisabled, 10 hearing impaired, 14 Down syndrome, and 13 developmentally delayed preschool children and their older, nondisabled siblings. Significant differences were found between the nondisabled and the disabled children. The siblings of disabled children appeared to be less conversationally sensitive with their brothers and sisters than the siblings of nondisabled children. Significant differences were also observed between the different types of disabilities. Few significant differences were found in variables believed to be facilitative of language development in the younger child, and those that were found tended not to be in the expected direction.

The bilingual child is seen as a unique source of information about the relation between input and intake. The strength of the association between language exposure estimates and vocabulary learning was examined for 25 simultaneous bilingual infants (ages 8 to 30 months) with differing patterns of exposure to the languages being learned. Using the MacArthur Communicative Development Inventories, standardized parent report forms in English and Spanish, the percentage of all words that were known in each language was calculated and then plotted against the estimates of language input (also in percentages). A significant correlation was found, r(25)= .82, p < .001. The correlation was also strong when examined point-by-point, even for children whose language environments changed by more than 20%; between observations, although it was not reliable at lower levels of exposure to Spanish. Especially for children with less input in the minority language, the factors which appeared to affect the strength of the association between input and amount learned in a language are discussed.

Two experiments were conducted with the purpose of investigating possible age effects on the abilities of older and younger adults to use contextual information to resolve ambiguous pronouns. In both experiments, subjects were presented with pairs of sentences (a leading sentence followed by a pronominal sentence) and were required to indicate the referent of the ambiguous pronoun. In both experiments, the older adults responded more slowly and were less accurate than the younger adults. However, both groups of subjects were equally influenced by the contextual information available, which was located in the leading sentence to aid in the resolution of the pronouns. Older adults did not demonstrate a specific impairment in the ability to use contextual information to resolve ambiguous pronouns. Nevertheless, agerelated difficulties in resolving pronouns may emerge, possibly as a function of an underspecified discourse model.

Six experiments examined the role of metaphorical knowledge in people's use and understanding of euphemisms and offensive expressions. Experiments 1 and 2 demonstrated that familiar euphemisms and dysphemisms are viewed as more appropriate and are easier to comprehend when there is a conceptual match between them and the context. The data from Experiments 3 and 4 showed a similar pattern for novel euphemisms and offensive phrases. Experiments 5 and 6 ruled out the hypothesis that the previous results were due to semantic priming. The findings from these experiments indicate that people's metaphorical conceptualization of a certain topic can influence the processing time and appropriate use of euphemistic and dysphemistic expressions.

The relationship between pragmatic referential communication skill and the cognitive ability to assess and assume another person's conceptual viewpoint was investigated in the autistic population. Ten high functioning autistic adolescents and young adults were matched for age and sex to normally developing controls and given referential communication and perspectivetaking tasks that had been previously demonstrated to be of comparable complexity. The groups were selected to be similar in terms of language skill.But despite their intact, elementary perspective-taking skills, the autistic subjects displayed significant communicative dysfunction. This suggests that factors other than a deficiency in the development of a “theory of mind” are significant contributors to the social communicative disorder associated with autism.

Transcriptions of 17 hours of recordings of babbling by fraternal twins in an English/Serbian language environment (1,454 utterances) were analyzed for basic aspects of articulatory organization, effects of the “twin situation,” and effects of the two ambient languages, English and Serbian. Predictions that babbling would be dominated by a “frame” provided by rhythmic mandibular oscillation were, for the most part, confirmed in the form of consonant-vowel co-occurrence constraints showing little active intersegmental tongue movement (one subject) and a predominance of “vertical” (mandible-induced) intersyllabic variegation (both subjects). A possible effect of the twin situation was observed in the form of unusually high frequencies of the consonants and vowels most frequent in babbling. The only prominent ambient language effect was a relatively high frequency of palatal glides (palatals are common in Serbian).

The purpose of this study was to investigate word reading abilities in first and third grade Spanish-speaking children who were learning to read in Spanish; the performance of skilled and less skilled readers was compared across measures that assessed phonological recoding ability, knowledge of grapheme-phoneme correspondences, and phonemic awareness. The findings suggest that Spanish-speaking children relied on phonological recoding strategies in the process of becoming readers. First grade, less skilled readers seemed to depend on partial letter-sound knowledge. Furthermore, spelling-sound correspondences appeared to be the main source of information used by first grade, skilled readers and third grade, less skilled readers. The latter seemed to lag behind skilled readers in the use of word-specific information. The phonemic awareness tasks displayed moderate to low correlations with reading ability in the less skilled groups. It is argued that the simple phonological structure of Spanish and its shallow orthography lead to the phonological processing of letter strings during reading acquisition.

This study compared vocal features of deception that can be measured by acoustic equipment with vocal features of deception that can be measured perceptually by human coders. As deception researchers have traditionally measured vocal behavior with either acoustic or perceptual methods (but not both), it is uncertain what correspondence, if any, exists between these methods. This study attempted to determine the degree of this correspondence. Deceptive interactions from an earlier study (Burgoon, Buller, Ebesu, & Rockwell, 1994) were used to conduct a detailed analysis of the vocal features of deceptive speech. The vocal samples were analyzed perceptually and acoustically. Results indicated moderate correlations between some acoustic and perceptual variables; neither measurement type, however, proved conclusively superior to the other in discriminating between truth and deception.

Preschoolers with specific language impairment (SLI) and two groups of children developing language normally (age-equivalent and MLU-equivalent) described motion and change-ofstate scenes while viewing a video program in a controlled experimental setting.Patterns of verb use were then analyzed across the three groups of children. Two findings were of particular interest. First, while the children with SLI relied heavily on the use of general all-purpose (GAP) verbs while describing the video scenes, the normally developing children produced an even higher percentage of CAP verbs. Second, the children with SLI made significantly more semantic errors in verb use. Unlike the normally developing children, the children with SLI were much more likely to produce a change-of-state verb. Hence, they produced cross-domain errors for motion-eliciting scenes (e.g., a moving figure was said to be “cracking” or “changing color”) and within-domain errors for change-of-state-eliciting scenes (e.g., a figure changing shape was said to be “changing to green”).

In the traditional gating technique, subjects hear increasing amounts of word-onset information from spoken words until the words can be correctly identified. The experiment reported here contrasted word-onset gating with results when words were gated from their word endings. A significant recognition advantage for words gated from their onsets was demonstrated. This effect was eliminated, however, when we took into account the number of word possibilities that shared overlapping phonology and the same stress pattern as the target words at their recognition points. These results support the position that the perceptual advantage of word-initial information can be understood within a general goodness-of-fit model of spoken word recognition.