TY - JOUR

T1 - Language Switching in Multilingual Reading

T2 - It's all about bigrams!

AU - Casaponsa, Aina

AU - Carreiras, Manuel

AU - Duñabeitia, Jon Andoni

PY - 2013

Y1 - 2013

N2 - How do multilinguals detect the language of a word that they are reading? Recent electrophysiological research has shown that bilingual readers identify the language in 200 milliseconds, as shown in masked priming experiments testing the switch cost effect. The main goal of this study is to understand the underlying mechanisms of this effect, investigating the extent to which multilinguals rely in sub-lexical orthographic regularities of the words to predict the language. In a first ERP experiment, we manipulated the bigrams of the words in the second language (Basque) as a function of their legality in the first language (Spanish). Spanish targets were preceded by unrelated words either in Spanish or Basque. Unrelated Basque words could contain bigram combinations that were either legal or illegal in the target language. Results were clear-cut, showing an astonishingly clear long-lasting switch cost effect in the N250 component for words in the non-target language containing illegal bigrams as measured in the target language, and a clear lack of such an effect for Basque words whose bigrams did also exist in Spanish. A second ERP experiment using the same language combination with a different set of materials and participants completely replicated these findings, and additional behavioral data extended them to single (unprimed) word reading. These pieces of evidence support the hypothesis that multilinguals rely on statistical regularities of the words in order to detect their language. Thus, basic orthographic features of the words guide multilingual language detection, and consequently multilingual visual word recognition processes.

AB - How do multilinguals detect the language of a word that they are reading? Recent electrophysiological research has shown that bilingual readers identify the language in 200 milliseconds, as shown in masked priming experiments testing the switch cost effect. The main goal of this study is to understand the underlying mechanisms of this effect, investigating the extent to which multilinguals rely in sub-lexical orthographic regularities of the words to predict the language. In a first ERP experiment, we manipulated the bigrams of the words in the second language (Basque) as a function of their legality in the first language (Spanish). Spanish targets were preceded by unrelated words either in Spanish or Basque. Unrelated Basque words could contain bigram combinations that were either legal or illegal in the target language. Results were clear-cut, showing an astonishingly clear long-lasting switch cost effect in the N250 component for words in the non-target language containing illegal bigrams as measured in the target language, and a clear lack of such an effect for Basque words whose bigrams did also exist in Spanish. A second ERP experiment using the same language combination with a different set of materials and participants completely replicated these findings, and additional behavioral data extended them to single (unprimed) word reading. These pieces of evidence support the hypothesis that multilinguals rely on statistical regularities of the words in order to detect their language. Thus, basic orthographic features of the words guide multilingual language detection, and consequently multilingual visual word recognition processes.

UR - https://publons.com/publon/11982353/

M3 - Meeting abstract

VL - 25

SP - 45

JO - Journal of Cognitive Neuroscience

JF - Journal of Cognitive Neuroscience

SN - 0898-929X

IS - Suppl.

ER -