ASHA journals
2 files

Children’s comprehension of object relative sentences (Ahmad Rusli & Montgomery, 2017)

Download all (1.5 MB)
posted on 2017-09-15, 21:32 authored by Yazmin Ahmad Rusli, James W. Montgomery
Purpose: The aim of this study was to determine whether extant language (lexical) knowledge or domain-general working memory is the better predictor of comprehension of object relative sentences for children with typical development. We hypothesized that extant language knowledge, not domain-general working memory, is the better predictor.
Method: Fifty-three children (ages 9–11 years) completed a word-level verbal working-memory task, indexing extant language (lexical) knowledge; an analog nonverbal working-memory task, representing domain-general working memory; and a hybrid sentence comprehension task incorporating elements of both agent selection and cross-modal picture-priming paradigms. Images of the agent and patient were displayed at the syntactic gap in the object relative sentences, and the children were asked to select the agent of the sentence.
Results: Results of general linear modeling revealed that extant language knowledge accounted for a unique 21.3% of variance in the children’s object relative sentence comprehension over and above age (8.3%). Domain-general working memory accounted for a nonsignificant 1.6% of variance.
Conclusions: We interpret the results to suggest that extant language knowledge and not domain-general working memory is a critically important contributor to children’s object relative sentence comprehension. Results support a connectionist view of the association between working memory and object relative sentence comprehension.

Supplemental Material S1. Experimental sentences.

Supplemental Material S2. Experimental nonverbal stimuli.

Ahmad Rusli, Y., & Montgomery, J. W. (2017). Children’s comprehension of object relative sentences: It’s extant language knowledge that matters, not domain-general working memory. Journal of Speech, Language, and Hearing Research. Advance online publication.


This study was funded by an Ohio University Student Enhancement Award to Yazmin Ahmad Rusli.