Our understanding of the relationships between lexical and phonological development has been enhanced in recent years by increased interest in this area from language scientists, psychologists and phonologists. This review article provides a summary of research, highlighting similarities and differences across studies. It is suggested that the research falls into two categories with different goals and different methodological approaches: (1) child-centered studies that examine the influences active in the prelinguistic and early-word period, emphasizing individual developmental patterns and the active role played by the child; and (2) studies inspired by research on word processing in adults; these focus on the effects of the phonological and lexical characteristics of the ambient language on underlying representations and word learning in children. The article concludes with suggestions for integrating the findings from the two approaches and for future research.

The present study investigated whether children's representations of morphosyntactic information are abstract enough to guide early verb learning. Using an infant-controlled habituation paradigm with a switch design, Japanese-speaking children aged 1 ; 8 were habituated to two different events in which an object was engaging in an action. Each event was paired with a novel word embedded in a single intransitive verb sentence frame. The results indicated that only 40% of the children were able to map a novel verb onto the action when the mapping task was complex. However, by simplifying the mapping task, 88% of the children succeeded in verb–action mapping. There were no differences in perceptual salience between the agent and action switches in the task. These results provide strong evidence that Japanese-speaking children aged 1 ; 8 are able to use an intransitive verb sentence frame to guide early verb learning unless the mapping task consumes too much of their cognitive resources.

Studies report that infants as young as 1 ; 3 to 1 ; 5 will seek out a novel object in response to hearing a novel label (e.g. Halberda, 2003; Markman, Wasow & Hansen, 2003). This behaviour is commonly known as the ‘mutual exclusivity’ response (Markman, 1989; 1990). However, evidence for mutual exclusivity does not imply that the infant has associated a novel label with a novel object. We used an intermodal preferential looking task to investigate whether infants aged 1 ; 4 could use mutual exclusivity to guide their association of novel labels with novel objects. The results show that infants can successfully map a novel label onto a novel object, provided that the novel label has no familiar phonological neighbours. Therefore, as early as 1 ; 4, infants can use mutual exclusivity to form novel word–object associations, although this process is constrained by the phonological novelty of a label.

Liaison and elision in French are phonological phenomena that apply across word boundaries. French-speaking children make errors in contexts where liaison/elision typically occurs in adult speech. In this study, we asked if acquisition of French liaison/elision can be explained in a constructivist framework. We tested if children's liaison/elision was sensitive to co-occurrence and meaning. We expected children's use of liaison/elision to correlate with their experience with French (estimated by vocabulary). Thirty-one French-speaking children (twenty-five bilingual) between three and five years old produced familiar vowel-initial words, following four words: (1) un, (2) deux, (3) un petit and (4) beaucoup de. The children with smaller French vocabularies produced many vowel-initial words and some consonant-initial chunks. The children with larger French vocabularies produced liaison/elision correctly across several frames while associating a number interpretation with liaised consonants. These results suggest that children use a variety of cues to construct the appropriate use of liaison/elision.

English-acquiring children frequently make pronoun case errors, while German-acquiring children rarely do. Nonetheless, German-acquiring children frequently make article case errors. It is proposed that when child-directed speech contains a high percentage of case-ambiguous forms, case errors are common in child language; when percentages are low, case errors are rare. Input to English and German children was analyzed for percentage of case-ambiguous personal pronouns on adult tiers of corpora from 24 English-acquiring and 24 German-acquiring children. Also analyzed for German was the percentage of case-ambiguous articles. Case-ambiguous pronouns averaged 63·3% in English, compared with 7·6% in German. The percentage of case-ambiguous articles in German was 77·0%. These percentages align with the children's errors reported in the literature. It appears children may be sensitive to levels of ambiguity such that low ambiguity may aid error-free acquisition, while high ambiguity may blind children to case distinctions, resulting in errors.

The experiments here build on the widely reported finding that children are most accurate when producing phonotactic sequences with high ambient-language frequency. What remains controversial is a description of the input that children must be tracking for this effect to arise. We present a series of experiments that compare two ambient-language properties, token and type frequency, as they contribute to phonotactic learning. Token frequency is the raw number of exposures children have to a particular pattern; type frequency refers to a count of abstract entities, such as unique words. Our results suggest that children's production accuracy is most sensitive to a combination of type and token frequency: children were able to generalize a target phonotactic sequence to a new word when familiarized with multiple word-types across tokens from multiple talkers, but not when presented with either word-types with no talker variability or multiple talker-tokens of a single word.

Children as young as two years of age are able to learn novel object labels through overhearing, even when distracted by an attractive toy (Akhtar, 2005). The present studies varied the information provided about novel objects and examined which elements (i.e. novel versus neutral information and labels versus facts) toddlers chose to monitor, and what type of information they were more likely to learn. In Study 1, participants learned only the novel label and the novel fact containing a novel label. In Study 2, only girls learned the novel label. Neither girls nor boys learned the novel fact. In both studies, analyses of children's gaze patterns suggest that children who learned the new information strategically oriented to the third-party conversation.

This study examined relationships between prelinguistic variables from the MacArthur-Bates CDI and the development of language comprehension and production in children with autism. Forty-four children were assessed at baseline and 6, 12, 24, 33 and 53 months later. Growth Curve Modeling was used to examine the extent to which three composite CDI variables and three CDI item groupings predicted language development over 4–5 years. When examined individually, prespeech and early gestures were significant predictors of change for both comprehension and production, but late gestures were not. In addition, initiating joint attention and games and routines predicted comprehension and production over 4–5 years, and conventional gestures also predicted production. When all factors were considered simultaneously, children's ability to participate in games and routines was the only significant predictor of language production over time. The results are discussed with regard to their implications for understanding the complex factors that affect developmental outcomes.

The acquisition of German participle inflection was investigated using spontaneous speech samples from six children between 1 ; 4 and 3 ; 8 and ten children between 1 ; 4 and 2 ; 10 recorded longitudinally at regular intervals. Child-directed speech was also analyzed. In adult and child speech weak participles were significantly more frequent than strong participles. Children's errors involved all elements of participle marking. All error types, including over-regularization, occurred from the beginning alongside correct forms. Errors decreased significantly over age. Over-regularization in the sense of -t affixation on strong verbs was significantly more frequent than erroneous -en suffixation on weak verbs but not than prefix and suffix omission. On participles with stem vowel change erroneous stem vowel was significantly more frequent than correct stem vowel with suffix error alone. Error patterns are explained in terms of frequencies, and participle inflection being learned as part of general verb inflection.

We applaud Stoel-Gammon's (this issue) call for a more comprehensive account of the relationship between lexicon and phonology, and we strongly endorse her suggestions for future research. However, we think that it will not be enough simply to integrate findings and methods from the adult-centered and child-centered literatures. Both of these literatures suggest that we need to rethink standard assumptions about what phonological representations are and how they emerge to support the very large vocabularies that speakers develop over the course of a lifetime. Our commentary focuses on three themes relevant to this reconceptualization.

Carol Stoel-Gammon has made a real contribution in bringing together two fields that are not generally jointly addressed. Like Stoel-Gammon, we have long focused on individual differences in phonological development (e.g. Vihman, Ferguson & Elbert, 1986; Vihman, Boysson-Bardies, Durand & Sundberg, 1994; Keren-Portnoy, Majorano & Vihman, 2008). And like her, we have been closely concerned with the relationship between lexical and phonological learning. Accordingly, we will focus our discussion on two areas covered by Stoel-Gammon (this issue) on which our current work may shed some additional light.

We investigate how Tamil- and Dutch-speaking adults and four- to five-year-old children use caused posture verbs (‘lay/stand a bottle on a table’) to label placement events in which objects are oriented vertically or horizontally. Tamil caused posture verbs consist of morphemes that individually label the causal and result subevents (nikka veyyii ‘make stand’; paDka veyyii ‘make lie’), occurring in situational and discourse contexts where object orientation is at issue. Dutch caused posture verbs are less semantically transparent: they are monomorphemic (zetten ‘set/stand’; leggen ‘lay’), often occurring in contexts where factors other than object orientation determine use. Caused posture verbs occur rarely in Tamil input corpora; in Dutch input, they are used frequently. Elicited production data reveal that Tamil four-year-olds use infrequent placement verbs appropriately whereas Dutch children use high-frequency placement verbs inappropriately even at age five. Semantic transparency exerts a stronger influence than input frequency in constraining children's verb meaning acquisition.

This study examined the effect of number of syllables and syllable structure on repetition of pseudo-words by Russian-speaking children with Specific Language Impairment (SLI) and typically developing (TD) children. One hundred and forty-four pseudo-words, varying in length and syllable complexity, were presented to two groups of children: 15 children with SLI, age range 4 ; 0 to 8 ; 8, and 15 TD children matched in age to the SLI group. The number of errors in the repetition of pseudo-words was analyzed in terms of the number of syllables and syllable complexity. The results demonstrated that children with SLI have deficits in working memory capacity. In addition to the pseudo-word length, the repetition performance was affected by syllable structure complexity.

This study tested the effect of the phoneme's linguistic affiliation (Standard Arabic versus Spoken Arabic) on phoneme recognition among five-year-old Arabic native speaking kindergarteners (N=60). Using a picture selection task of words beginning with the same phoneme, and through careful manipulation of the phonological properties of target phonemes and distractors, the study showed that children's recognition of Standard phonemes was poorer than that of Spoken phonemes. This finding was interpreted as indicating a deficiency in the phonological representations of Standard words. Next, the study tested two hypotheses regarding the specific consequences of under-specified phonological representations: phonological encoding versus phonological processing. These hypotheses were addressed through an analysis of the relative power of distractors. The findings revealed that children's difficulty in accessing Standard Arabic phonemes was due to a difficulty in the phonological encoding of Standard words. We discuss the implications of the findings for language and literacy development in diglossic Arabic.

The study examines strategies multilingual children use to interpret grammatical relations, focusing on their two primary languages, Lajamanu Warlpiri and Light Warlpiri. Both languages use mixed systems for indicating grammatical relations. In both languages ergative–absolutive case-marking indicates core arguments, but to different extents in each language. In Lajamanu Warlpiri, pronominal clitics in a nominative–accusative pattern also indicate core arguments, and in Light Warlpiri word order in a nominative–accusative pattern partially does so. The study asks which sentence interpretation strategies children rely on most, when they learn to rely on them and whether cross-linguistic influences are seen. Children aged 5 ; 0, 7 ; 0 and 9 ; 0 and adults saw paired, animated events simultaneously on video and heard a transitive sentence spoken. The participants pointed to the event depicted by the sentence heard. Adults used a case-marking strategy consistently in both languages. Children initially used both case-marking and word order strategies, but used case-marking more often as age increased.

This study investigates the acquisition of word-initial Korean obstruents (i.e. stops, affricates and fricatives). Korean obstruents are characterized by a three-way contrast among stops and affricates (i.e. fortis, aspirated and lenis) and a two-way fricative contrast (i.e. fortis and lenis). All these obstruents are voiceless word-initially. Cross-sectional data were collected from forty Korean children aged 2 ; 6 (year;month), 3 ; 0, 3 ; 6 and 4 ; 0, and the acquisition patterns of Korean obstruents were explored based on productions of mono- and multisyllabic words. Results confirm the universal patterns: stops were acquired before affricates and fricatives. In terms of order of acquisition across different laryngeal types, lenis stops were the last to be acquired. For Korean fricatives, in contrast, the lenis category appears to be acquired earlier than the fortis category. This study proposes that this order of acquisition may be explained by articulatory complexity associated with language-specific phonetic properties.

This study tracked the order in which ten beginning spellers (M age=5 ; 05; SD=0·21 years) mastered the correct spellings of common inflectional suffixes in English. Spellings from children's journals from kindergarten and grade 1 were coded. An inflectional suffix was judged to be mastered when children spelled it accurately in 90 percent of the contexts in which it was grammatically required, a criterion used to study the order of acquisition of grammatical morphemes in oral language. The results indicated that the order in which children learned to spell inflectional suffixes correctly is similar to the order in which they learn to use them in oral language, before school age. Discrepancies between the order of mastery for inflectional suffixes in written and oral language are discussed in terms of English spelling conventions, which introduce variables into the spelling of inflected words that are not present in oral language.

This study examined the extent and nature of lexical tone deficit in Chinese developmental dyslexia. Twenty Cantonese-speaking Chinese dyslexic children (mean age 8 ; 11) were compared to twenty average readers of the same age (CA control group, mean age 8 ; 11), and another twenty younger average readers of the same word reading level (RL control group, mean age 7 ; 4) on different measures of lexical tone awareness, rhyme awareness and visual–verbal paired-associate learning. Results showed that the Chinese dyslexic children performed significantly worse than the CA but not the RL control groups in nearly all the lexical tone and rhyme awareness measures. Analyses of individual performance demonstrated that over one-third of the dyslexic children showed a deficit in some aspects of tone awareness. Tone discrimination and tone production were found to correlate significantly with Chinese word reading. These findings confirm that Chinese dyslexic children show weaknesses in tone awareness.

When Roger Brown selected Adam, Eve and Sarah to be the first three participants in the modern study of child language, one of the criteria was the intelligibility of their speech (Brown, 1973). According to the prevailing view at the time, accuracy of pronunciation was a peripheral phenomenon that had nothing to do with the development of language qua language. So why not study children who were easy to transcribe? One reason why not, according to Stoel-Gammon (SG; this issue), is that the difficulty of accurately producing sounds influences the words children acquire and the rate at which they acquire them. (It's true that Roger Brown's focus was on the child's acquisition of morphosyntax, but articulation was assumed to be peripheral to everything back then.) This interaction between the articulatory skill of children and phonological properties of words is just one of the mutual influences between phonology and the lexicon SG describes. In her target article, SG brings together data from a wide range of investigations to build an account of how phonology and the lexicon interact in development.

Children with specific language impairment (SLI) vary widely in their ability to use tense/agreement inflections depending on the type of language being acquired, a fact that current accounts of SLI have tried to explain. Finnish provides an important test case for these accounts because: (1) verbs in the first and second person permit null subjects whereas verbs in the third person do not; and (2) tense and agreement inflections are agglutinating and thus one type of inflection can appear without the other. Probes were used to compare the verb inflection use of Finnish-speaking children with SLI, and both age-matched and younger typically developing children. The children with SLI were less accurate, and the pattern of their errors did not match predictions based on current accounts of SLI. It appears that children with SLI have difficulty learning complex verb inflection paradigms apart from any problem specific to tense and agreement.