Do Bilinguals Automatically Activate Their Native Language When They Are Not Using It?

Area of Science:

• Psycholinguistics
• Bilingualism research
• Cognitive science

Background:

• Current models of lexical access in bilinguals often assume simultaneous activation of both languages.
• Evidence for this includes the automatic activation of translation equivalents in the non-target language.
• This phenomenon is frequently cited to support the co-activation hypothesis.

Purpose of the Study:

• To propose an alternative account for findings attributed to automatic translation equivalent activation in bilinguals.
• To challenge the notion that bilinguals always activate both languages concurrently.
• To present a model explaining observed data without invoking cross-language activation.

Main Methods:

• Development of a computational model of bilingual language learning.
• The model focuses on the mapping of phonological relationships from the native to the non-native language.
• Simulation of language processing to test explanatory power against existing findings.

Main Results:

• The proposed model successfully explains findings previously interpreted as evidence for automatic translation equivalent activation.
• The model demonstrates how native language structure transfer can account for these observations.
• Results suggest that cross-language activation may not be a necessary component of bilingual lexical access.

Conclusions:

• Observed phenomena in bilinguals are compatible with a language learning account based on structural transfer, not necessarily translation equivalent activation.
• Bilinguals may learn new languages by mapping existing phonological structures, rather than through constant cross-linguistic activation.
• This offers a parsimonious explanation for key findings in bilingual language processing research.