Several descriptions of the transition from single to multiword utterances use prosody as an important diagnostic criterion. For example, in contrast to successive single-word utterances, ‘real’ two-word utterances are supposed to be characterized by a unifying intonation contour and a lack of an intervening pause. Research on the acquisition of prosody, however, revealed that control of the phonetic parameters pitch, loudness, and duration is far from complete at such an early stage. In this study, we examine the interaction between the development of different types of syntactic structures and their prosodic organization. Data from a detailed production record of a monolingual German-learning boy is analysed both auditorily and acoustically with a focus on four different types of two-word utterances produced between 2;0 and 2;3. Two major findings are reported here. First, the different types of two-word utterances undergo individual trajectories of prosodic (re-)organization, in part depending on the time course in which they become productive. This suggests that different types of multiword utterances become prosodically fluent at different points in time. Second, the variability of prosodic features such as pauses and stress pattern is very high at the onset of combinatorial speech. Consequently, fluency or disfluency of individual examples should not be used as a reliable criterion for their syntactic status and we recommend caution when taking prosody as a cue for syntactic development.

The study explored early syntactic development, and tested the hypothesis that children use similarity of meaning in order to move beyond the learning of individual item-based multiword constructions. The first 6 types of verb–object (VO) constructions in Hebrew-speaking children were analysed for the occurrence of transfer of learning and facilitation, as well as for the semantic similarity of the direct objects (DO). Longitudinal naturalistic speech corpora of 20 children (1;06–2;06) were analysed. We found facilitation (increased rate of learning) among the first 6 types of VO constructions (each type built on a different verb) as evidenced by the accelerating growth curves. Next, we measured the semantic similarity of the DOs using an 8-category system including Patient, Theme and Object of Result. The first 6 DO types represented 3·95 different semantic roles. On the average, after the first VO construction was learned, 3 out of the following 5 constructions produced were not preceded by another VO construction where the DO was of the same semantic category. The results indicate that facilitation of learning of early syntax is most probably NOT mediated by semantic similarity.

Predictive relations were examined between measures of 20 mothers' behavioural and verbal general and specific responsiveness and intrusive and supportive directiveness and their children's subsequent expressive vocabularies during three developmental periods with endpoints at the beginning, middle, and end of the second year: 0;10 to 1;1, 1;1 to 1;5, and 1;5 to 1;9. Regression analyses, controlling for mothers' utterance frequencies and children's initial lexicons, revealed considerable consistency between reported and observed lexicons but changing patterns of predictive relations with development. During the first period, behavioural, but not verbal, measures of maternal responsiveness and supportive directiveness were positively predictive. In period two, verbal, but not behavioural, measures predicted children's vocabularies, with specific responsiveness and supportive directiveness as positive predictors and intrusive directiveness as a negative predictor. During the final period, mothers' behavioural and verbal responsiveness and behavioural supportive directiveness positively predicted and their verbal intrusive directiveness negatively predicted children's lexical growth.

This research addresses the question of why some children are disposed to making a large number of pronoun case errors and others are not. The answer proposed is that when pronoun paradigm building outstrips the development of INFL, children become especially vulnerable to erring in the choice of pronominal word form, resulting in pronoun case error. On the other hand, when pronoun paradigm building proceeds more conservatively, the risk of error is reduced. The spontaneous sentence production of children observed in naturalistic caregiver – child interaction from a cross-section of 44 children (2;0–4;0) is used to support this proposal. The data show that pronoun case error was minimal among children who had strong INFL. However, among children with weak INFL there was a wide range of variation, some children making many errors and others making none. Analysis of variance confirmed that this variation was strongly related to the dispersion of production attempts across an extended pronoun paradigm, such that, the fewer cells attempted, the lower the error rate. These findings show that pronoun case errors are not an inevitable result of grammatical development, but may conceivably be avoided altogether if paradigm building proceeds at a rate commensurate with the child's development of INFL.

Three experiments investigated the ability of eight-year old children with poor language comprehension to produce past tense forms of verbs. Twenty children selected as poor comprehenders were compared to 20 age-matched control children. Although the poor comprehenders performed less well than controls on a range of tasks considered to tap verbal-semantic abilities, the two groups showed equivalent phonological skills. Poor comprehenders performed as well as control children when asked to inflect novel verbs and regular verbs. In contrast, poor comprehenders were less skilled than controls at inflecting both high frequency and low frequency irregular verbs. Although the predominant error pattern for all children was to over-regularize, this was most marked in the poor comprehenders; control children were more likely to produce errors that contained knowledge of the irregular form than poor comprehenders. In addition, the ability to inflect irregular verbs was related to individual differences in verbal-semantic skills. These findings are discussed within a framework in which verb inflection is related to underlying language skills in both the phonological and semantic domains.

The family size of the constituents of compound words, or the number of compounds sharing the constituents, has been shown to affect adults' access to compound words in the mental lexicon. The present study was designed to see if family size would affect children's segmentation of compounds. Twenty-five English-speaking children between 3;7 and 5;9 were asked to explain the meaning of existing compounds with constituents of varying family size to an alien puppet. The results showed that children were more likely to mention the modifier of compounds if they came from large constituent families than if they came from small constituent families. Other variables were also shown to have some, but smaller effects on children's parsing, including the frequency of the constituent words and the compounds, whether the compounds were already known, and age. These results suggest that children's segmentation of compounds might be facilitated by analogy with other compounds already in their vocabularies.

The specificity of infants' phonological representations is examined by comparing their sensitivity to mispronunciations of novel and familiar words, using the preferential looking task. 29 children at 1;2 were trained and tested on familiar and novel word–object pairs. Children showed evidence of sensitivity to mispronunciations of novel and familiar words, indicating detailed phonological representations. Discrepancies between this study and earlier investigations are discussed with reference to differences between habituation and preferential looking tasks.

This article explores whether infants are able to learn words as rapidly as has been reported for preschoolers. Sixty-four infants aged 1;6 were taught labels for either two moving images or two still images. Each image–label pair was presented three times, after which comprehension was assessed using an adaptation of the intermodal preferential looking paradigm. Three repetitions of each label were found to be sufficient for learning to occur, fewer than has previously been reported for infants under two years. Moreover, contrary to a previous finding, learning was equally rapid for infants who were taught labels for moving versus still images. The findings indicate that infants in the early stages of acquiring a vocabulary learn new word-referent associations with ease, and that the learning conditions that allow such learning are less restricted that was previously believed.

This study charts the development of the genitive masculine inflection, one of the most irregular parts of the Polish case-marking system. 72 Polish children aged from 2;3 to 10;8 participated in a nonce word production experiment testing their ability to supply the genitive form and their sensitivity to the semantic factors determining the choice of ending. Results indicate that productivity, or the ability to supply the inflected form of some nonce words, emerges early: 78% of the two-year-olds were able to inflect at least one test item. However, mastery, or the ability to consistently supply the correct ending, takes considerably longer to develop, and adultlike levels of provision are not reached until about age 10;0.

We investigated whether manipulating the perceived novelty of nameless objects would influence two-year-olds' tendency to map novel words to these objects. In Experiment 1, children who had been pre-exposed to target nameless objects were more likely to map novel words onto those objects than children who were not pre-exposed to the objects or children who were pre-exposed to non-target members of the nameless object categories. In Experiment 2, children who were pre-exposed to a nameless object were more likely to assign the novel label to that object than to either a familiar object or an unfamiliar object that had not been pre-exposed. The results of these studies suggest that reducing the novelty of nameless objects increases two-year-olds' tendency to map a novel word to a nameless object.

Two experiments examined the relation between early object name learning and the ability to represent objects by their abstract shapes. In Experiment 1, two-year-old children with productive vocabularies in the bottom 20th percentile – ‘late talkers' – were compared with (1) same-age children with larger vocabularies, and (2) younger children matched for productive vocabulary, on their ability to recognize named common objects. Object categories were represented two ways: by lifelike, perceptually rich toys, and by grey caricatures of those objects' abstract shapes. All 3 groups recognized lifelike objects equally well. Both typically-developing control groups were better than late talkers at recognizing shape caricatures of objects whose names they knew. In Experiment 2, late talkers and age-matched controls identified named objects represented by lifelike toys and by duplicates of those toys covered in grey textured paint. Age-matched controls knew more of the object names overall, but both they and the late talkers performed equally well on both kinds of test objects. Thus, late talkers had some difficulty in Experiment 1 recognizing objects from abstract shape cues, but no difficulty in Experiment 2 when the shape cues were realistic. The findings imply a relation between the growth of productive vocabulary and the emergence of the ability to represent object categories by abstract shape.