Cross-Linguistic Similarity and Task Demands in Japanese-English Bilingual Processing

David B Allen,Kathy Conklin,Kevin Paterson

doi:10.1371/journal.pone.0072631

Abstract

Even in languages that do not share script, bilinguals process cognates faster than matched noncognates in a range of tasks. The current research more fully explores what underpins the cognate ‘advantage’ in different script bilinguals (Japanese-English). To do this, instead of the more traditional binary cognate/noncognate distinction, the current study uses continuous measures of phonological and semantic overlap, L2 (second language) proficiency and lexical variables (e.g., frequency). An L2 picture naming (Experiment 1) revealed a significant interaction between phonological and semantic similarity and demonstrates that degree of overlap modulates naming times. In lexical decision (Experiment 2), increased phonological similarity (e.g., bus/basu/vs. radio/rajio/) lead to faster response times. Interestingly, increased semantic similarity slowed response times in lexical decision. The studies also indicate how L2 proficiency and lexical variables modulate L2 word processing. These findings are explained in terms of current models of bilingual lexical processing.

Highlights

IntroductionThere is considerable evidence that cognates are processed more quickly than matched noncognates in a range of production (word naming: [1]; picture naming: [2,3]; word translation: [4,5,6,7]) and comprehension tasks (lexical decision: [8,9]; masked priming: [10,11,12,13,14]; progressive de-masking: [9]; sentence comprehension: [15,16])
There is considerable evidence that cognates are processed more quickly than matched noncognates in a range of production and comprehension tasks
The present study extends previous research [20], in which a cognate effect was found in L2 picture naming in different script bilinguals, by utilising continuous measures of similarity as well as by accounting directly for other factors in a mixed-effects model

Summary

Introduction

There is considerable evidence that cognates are processed more quickly than matched noncognates in a range of production (word naming: [1]; picture naming: [2,3]; word translation: [4,5,6,7]) and comprehension tasks (lexical decision: [8,9]; masked priming: [10,11,12,13,14]; progressive de-masking: [9]; sentence comprehension: [15,16]). The robustness of this cognate facilitation effect is attested across a wide range of tasks and with a number of first and second languages. The cognate advantage has been found even when languages do not share a script (e.g., Japanese-English, Korean-English, Hebrew-English, Greek-French). Cognates share meaning (semantics; S) and form (phonological and/or orthographic; P and O) across languages. Their processing advantage could be underpinned by overlap in S, P, and/or O. The description of cognates in the psycholinguistic literature is usually based on the degree of overlap of O/P and S features across languages, instead of being described etymologically. This work has been done in languages that share a script, which means that the contribution of O and P overlap is hard to disentangle

Objectives

Methods

Results

Conclusion