Antecedent priming at trace positions in children's sentence processing

Roberts, L., Marinis, T. and Clahsen, H. (2007) Antecedent priming at trace positions in children's sentence processing. Journal of Psycholinguistic Research, 36 (2). pp. 175-188. ISSN 0090-6905 doi: 10.1007/s10936-006-9038-3

Abstract/Summary

The present study examines whether children reactivate a moved constituent at its gap position and how children's more limited working memory span affects the way they process filler-gap dependencies. 46 5-7 year-old children and 54 adult controls participated in a cross-modal picture priming experiment and underwent a standardized working memory test. The results revealed a statistically significant interaction between the participants' working memory span and antecedent reactivation: High-span children (n = 19) and high-span adults (n = 22) showed evidence of antecedent priming at the gap site, while for low-span children and adults, there was no such effect. The antecedent priming effect in the high-span participants indicates that in both children and adults, dislocated arguments access their antecedents at gap positions. The absence of an antecedent reactivation effect in the low-span participants could mean that these participants required more time to integrate the dislocated constituent and reactivated the filler later during the sentence.