Individual Submission Summary
Share...

Direct link:

Poster #112 - Examination of Monolingual (English) and Bilingual (English/Spanish; English/Urdu) Children’s Syntactic Awareness

Fri, March 22, 12:45 to 2:00pm, Baltimore Convention Center, Floor: Level 1, Exhibit Hall B

Integrative Statement

The current study examined syntactic awareness for monolingual and bilingual five- and six-year-olds. Syntactic awareness, or awareness of grammaticality, was examined in two structural types (i.e., gender representation, word order). It was predicted that cross-linguistic influence would affect bilingual children’s performance due to overlap in subject-object-verb word order and determiner forms for gender labels between languages. Based on the metalinguistic hypothesis, expected differences may only be present at high levels of receptive vocabulary (Nagy, 2007). Findings may also be based on task characteristics, such as whether the presented sentences were correct or incorrect.

Participants included 121 children: 62 monolingual English-speaking children (Mage=6;1, range: 5;7–6;2), 35 bilinguals who spoke English/Spanish (Mage=6;0, range: 5;6–6;1), and 24 bilinguals who spoke English/Urdu (Mage=6;0, range: 5;7–6;1). Children were given the English Peabody Picture Vocabulary Test-III, a heritage language version if bilingual, and a 40-item English grammaticality judgment task containing correct and incorrect sentences. Within gender representation sentences, agrammatical sentences versus ambiguous agreement differences were explored (see Table 1). Within word order sentences, strict word order violations versus semantic anomaly differences were investigated.

There were no bilingual advantages, which is in line with studies showing inconsistent findings when engagement of executive functions is necessary (Barac & Bialystok, 2012). When regression analyses were performed, partial support was found for the interplay between receptive vocabulary and syntactic awareness. English vocabulary was a positive predictor of performance for the bilingual children (see Table 2). However, vocabulary ability in their heritage language did not predict performance, suggesting that the connection between receptive vocabulary and syntactic awareness may only exist within a language.

Receptive vocabulary and correctness of the presented sentences interacted with performance across the groups. Overall, gender violations and correct sentences were easier to identify than word order violations or incorrect sentences for all children. In the low- and middle-vocabulary ranges, children were better at identifying correct than incorrect constructions and were better at detecting correct and incorrect word order than gender violations, t(23-60) > 4.83, p < .0001. In the high-vocabulary range, children were equally as good at identifying correct and incorrect word order violations at a high accuracy level, they were better at identifying correct than incorrect gender constructions, and they were better at detecting incorrect word order constructions than incorrect gender constructions, t(35) > -5.33, p < .002.

In fact, those with high vocabulary scores (90% accuracy) were more accurate than the other groups (49% accuracy) when identifying agrammatical and asemantical word order constructions, which could be incorrect either by the misplacement of the subject, object, and verb or because of a semantic anomaly created by the word order, t(58-95) > -2.83, p < .006. In terms of gender violations, children in the high-vocabulary range were better at detecting incorrect gender agreement than the other groups, t(58-95) > -3.21, p < .002. No differences were found between the vocabulary groups when identifying ambiguous gender violations.

These results will be discussed in the context of the metalinguistic hypothesis and cross-linguistic influence for bilingual children.

Authors