ABSTRACT
This study examines the relation between lexical and phonological variables in 40 French-speaking children, aged 2;5. Specifically, it examines the influence of phonetic complexity, phonological production, phonological memory and neighbourhood density (ND) on vocabulary size. Children were divided into four groups on the basis of their scores on the French version of the Communicative Developmental Inventory (CDI): late1 (< 10%ile), late2 (15–25%ile), middle (40–60%ile) and precocious (> 90%ile). The children’s lexicons were coded in terms of phonetic complexity and ND (one-and two-syllable words), and their production capacities were determined from measuring percent consonants correct (PCC) and the number of syllable-initial (CSI) and -final (CSF) consonants in their phonetic inventories. The children also took part in a non-word repetition (NWR) task. Results indicated significant group differences in all four sets of variables. Children with larger vocabularies selected words with greater phonetic complexity and with lower ND values. They had superior PCC, CSI and NWR scores compared to children with smaller vocabularies. Linear regression analyses indicated that 76% of variance in vocabulary size could be accounted for by ND in combination with phonetic complexity and CSI. Our findings are consistent with previous studies which show that ND plays an important role in accounting for variance in vocabulary size. They also indicate that phonetic complexity and phonological production influence lexical acquisition.
Acknowledgements
I would like to thank Christophe dos Santos and Sophie Kern for their advice and helpful discussion. I would also like to thank the following students for their assistance in phonetic transcription and data-coding: Lucile da Silva, Elisa Chaplin and Pauline Mudry.
Declaration of interest
The authors report no conflicts of interest.
Notes
1 It is well acknowledged that a NWR task may tap many phonological skills apart from phonological memory including speech perception, phonological representation and articulatory abilities (see Coady & Evans, Citation2008). It is most commonly known as a measure of phonological memory which is how we refer to it in the current study.
2 Unlike Stokes et al. (Citation2012b), we do not include two syllable words such as “p(e)tit” and “ch(e)val” in the analyses of ND for one-syllable words. Although these words may be realized as monosyllabic, research by Andreassen (Citation2013) shows that the variant containing schwa is more frequent in children’s productions as well as in the input to children. Thus, these words have been coded as disyllabic.
3 One reason for the lack of significant differences in the late groups was the reduced numbers of words in their lexicons. Child 66 (late1), for example, obtained the highest phonetic complexity value (i.e. 4.07) of all children but his score was based on only 15 items, several of which were high phonetic complexity words (e.g. camion de pompier, glace, chocolat, popcorn).
4 We use the total IFDC score as our dependent variable of vocabulary size. Preliminary analyses revealed that similar correlations were obtained regardless of whether we took the total IFDC or the IFDC-R as dependent variable.
5 The Variance Inflation Factors (VIF) were 2.30 for ND1, 1.32 for phonetic complexity and 1.9 for CSI suggesting acceptable levels for multi-collinearity.