Can selectively activating language optimize multilinguals' performance?

; (). Can selectively activating language optimize multilinguals' performance?. In: 5th International Conference of 3rd Language Acquisition and Multilingualism. University of Stirling, 3.-5. September. Stirling: University of Stirling.

The paper reports on the results of experiments designed to test Grosjean`s language mode model (2001) with translation students in Switzerland proficient in at least three languages (i.e. German, English, Spanish, French and/or Italian). Although some research has referred to language mode with respect to lexical processing in multilinguals (e.g. Dijkstra & van Hell 2003; Marian & Spivey 2003), little has been directed to verifying the model for translation processes, where two languages are necessarily involved. Grosjean?s model makes no clear predictions about whether translators should have the target language highly activated before and during the translation process (to optimally produce the target text), or the source language (to optimally understand and therefore translate the source text), or both, and in that case to what degree the two languages should be activated relative to each other. Preliminary results from a pilot study (reported in Ehrensberger-Dow & Jekat 2005) suggest that activating the source language of a translation version (i.e. by having students engage in discussions in Spanish before attempting a Spanish-German translation) can have a positive effect on translation performance. However, those results were based on a small sample of students translating into their dominant language, so it was not clear whether the status of the language with respect to the translation version (source or target) or in the individual translator (dominant or non-dominant) was important.

The research question of interest here is whether language separation (and presumably avoidance of negative transfer) can be fostered by selectively activating one of the working languages in the translation process. It was addressed by a series of experiments that controlled for activation of various languages and comparisons of multilinguals? performance in different tasks and conditions. The experiments included tasks such as written translations of texts into German, Spanish or English, sight translations (unprepared, oral) of written texts and judgments of translation variants. Findings indicate that type of language activation and duration in relation to the task at hand might have differential effects on linguistic performance that are related more to the status of the languages for the multilingual individuals concerned than to whether the language is the source or target language of a translation version. The results contribute to an increased understanding of multilingual processing and the multilingual lexicon as well as potential optimization of multilingual education, including that of translators and interpreters.