It’s all in the eyes: How language dominance, salience, and context affect eye movements during idiomatic language processing

Abstract

This paper reports an eye movement study and the effects of salience, context, and language dominance on the processing of idiomatic expressions by Spanish–English bilinguals. Salient meanings of figurative expressions are those which are processed first and accessed automatically from the mental lexicon, regardless of contextual bias (Giora 2003). The research conducted so far with second language (L2) learners and bilingual participants has shown that the literal meaning of L2 idioms might be more salient than the figurative one in the course of their processing by non-native language users (e.g. Kecskes 2000; Liontas 2002; Cieślicka 2006; Cieślicka and Heredia 2011). In addition, research findings suggest that the degree of language dominance, or which language is more readily accessible due to usage (Heredia 1997; Heredia and Altarriba 2001; Altarriba and Basnight-Brown 2007), might be a factor in bilingual processing. To investigate whether the degree of literal and figurative activation in bilingual idiom processing may be modulated by language dominance (i.e. dominant vs. nondominant), we recorded eye movements of Spanish–English bilinguals, dominant either in Spanish or in English, while they were reading ambiguous (literally plausible, such as ‘kick the bucket’) English idioms. Each idiom was used either in its figurative or literal meaning and embedded in a sentence with neutral preceding context, in which case its figurative (‘Within seconds she realized she was in deep water, and that she would very soon come to regret her words’) or literal (‘Within seconds she realized she was in deep water, and that she would very soon have to swim back towards the shore’) meaning became clear due to the subsequent disambiguating information, or the preceding supportive context clearly biasing one of the meanings (e.g. figurative biased: ‘Since both of us were equally guilty of causing the overspend, we both knew we were in deep water, and very likely to lose our jobs’). As predicted, the results indicated that the effects of salience and context on eye movement patterns are modulated by language dominance.