Perception and production of inflections 1 Sentence - position effects on children ’ s perception and production of English 3 rd person singular – s

Purpose: Two-year-olds produce 3 person singular –s more accurately on verbs in sentencefinal compared to sentence-medial position. This study was designed to determine whether these sentence position effects can be explained by perceptual factors. Methods: For this purpose, we compared 22and 27-month-olds’ perception and elicited production of 3 person singular –s in sentence medial versus final position. Perception was assessed by measuring looking/listening times to a one-screen display of a cartoon paired with a grammatical vs. ungrammatical sentence (e.g., She eats now vs. She eat now). Results: Children at both ages demonstrated sensitivity to the presence/absence of this inflectional morpheme in sentence-final, but not sentence-medial position. Children were also more accurate at producing 3 person singular –s sentence-finally, and production accuracy was predicted by vocabulary measures as well as by performance on the perception task. Conclusions: These results indicate that children’s more accurate production of 3 person singular –s in sentence-final position cannot be explained by articulatory factors alone, but that perceptual factors play an important role in accounting for early patterns of production. The findings also indicate that perception and production of inflectional morphemes may be more closely related that previously thought. Perception and production of inflections 3 Introduction Children’s production of closed-class grammatical items (e.g., function words such as the, and, of, and inflectional morphemes such as -ing, -ed, -ly, -s) emerges later than their production of open-class, content words, and continues to be variable during early acquisition (Bloom, 1970; Brown, 1973). This variability in the acquisition of closed-class items is systematic. For example, longitudinal data from 1-3-year-olds and cross-sectional data from 2-year-olds shows that children produce 3 person singular –s more accurately sentence-finally compared to sentencemedially even when mean length of utterance (MLU), utterance length, and final syllable structure complexity of the inflected verb were controlled (Song, Sundara, & Demuth, 2009). Other studies have also found sentence-position effects on the production of grammatical morphemes by children with language impairment (cf. Dalal & Loeb, 2005; Norbury, Bishop & Briscoe, 2001; Leonard, Miller & Owen, 2000). One possible explanation for these findings concerns the greater articulatory/planning complexity of producing verbs in sentence-medial compared to sentence-final position. It has been shown that children’s productions begin to exhibit longer duration at phrase boundaries around the time they begin to produce word combinations (Snow, 1994, 1998). This is likely to give children more time to produce coda consonants and morphemes sentence-finally as compared to sentence-medially (cf. Kirk & Demuth, 2006). Inflectional morphemes like the 3 person singular -s might also be more challenging in utterance-medial position due to the fact that another word follows, necessitating the planning of additional articulatory gestures. In contrast, at the end of an utterance, no additional gestural planning is immediately required (Song et al., 2009). Such an articulatory/planning complexity account of sentence-position Perception and production of inflections 4 effects predicts that children’s production, but not their perception of 3 person singular –s should be sensitive to sentence-position effects. However, an alternate explanation of these results emerges when prosodic factors involving distributional/durational issues are considered. In an examination of conversation with and stories for children, Hsieh et al., (Hsieh, Leonard & Swanson, 1999) found that 52% of nouns with plural –s occurred in sentence-final position, whereas only 16% of verbs with 3 person singular –s occur sentence-finally. Given the fact that utterance-final syllables are lengthened in English, the average duration of the 3 person singular –s is about 25% shorter than that of the plural –s. Thus, it is also possible that the shorter duration of medial 3 person singular –s itself leads to less than perfect perception and production of this morpheme. The goal of this study was therefore to further investigate why English-learning children produce 3 person singular –s more accurately sentence-finally, and the extent to which this might be influenced by perceptual versus articulatory/planning factors. The results could help shed light on factors affecting the acquisition of grammatical morphemes more generally. In the following sections, we first review the literature on the perception of closed-class functional elements during acquisition, then we discuss issues relating to the use of preference as an index of