Online processing of causal relations in beginning first and second language readers

Ludo Verhoeven,Liza J Van Den Bosch,Eliane Segers

doi:10.1016/j.lindif.2017.11.007

Ludo Verhoeven, Liza J Van Den Bosch + Show 1 more

Open Access

https://doi.org/10.1016/j.lindif.2017.11.007

Copy DOI

Abstract

We investigated online processing of causal relations in beginning first (L1) and second language (L2) readers (8–10years old). By means of eye-tracking, we measured children's processing times of two-clause sentences including a causal relation. Two text-related factors were investigated: coherence marking (i.e., presence vs. absence of the Dutch connective omdat ‘because’) and linear order of clauses (i.e., cause-effect vs. effect-cause). In addition, syntactic knowledge was included as a child-related factor of interest. The results showed that coherence marking and individual differences in syntactic knowledge influenced children's online sentence processing. In contrast to L1 readers, the absence of a connective led to longer sentence processing times for L2 readers with lower syntactic knowledge; they experienced more difficulty with processing sentences in which no connective was present. Apparently, L2 readers with limited syntactic knowledge benefit from coherence marking provided by a connective, which allows them to establish the causal coherence relation between clauses in a more efficient way. Reversing the linear order of clauses did not affect children's online sentence processing. This study provides an initial step towards the use of online measures to examine sentence processing in beginning L1 and L2 readers aimed at gaining more insight into L2 reading comprehension difficulties.

Highlights

The following research question was central to the present study: To what extent are beginning L1 and L2 readers' online processing times of sentences with causal relations influenced by (a) coherence marking, (b) linear order of clauses, and (c) individual differences in syntactic knowledge? For coherence marking, we expected that beginning readers would benefit from the presence of connectives, and that L2
We examined online sentence processing in third-grade monolingual Dutch children (L1 readers) and bilingual Turkish-Dutch children (L2 readers)
We investigated to what extent their online processing times of sentences with causal relations were influenced by (a) coherence marking, (b) linear order of clauses, and (c) individual differences in syntactic knowledge

Summary

Introduction

Children who are confronted with the challenging task of learning to read in their second language, such as children from language minorities for whom the language of instruction at school differs from their home language, often experience reading comprehension difficulties (Droop & Verhoeven, 2003; Lesaux, Lipka, & Siegel, 2006; Mancilla-Martinez & Lesaux, 2010). We collected children's scores on two standardized tests for vocabulary knowledge and reading comprehension, which are paper-and-pencil tasks including seventy and fifty multiple-choice question, respectively. These tests are administered yearly in class by the teacher as part of the national Dutch student-monitoring system in primary schools developed by the Dutch National Institute for Educational Measurement (CITO).

Objectives

Methods

Results

Conclusion