Portuguese-english interlingual homophones in word reading and translation

Abstract

This study aimed at contributing to the understanding of bilingual lexical access in the area of Psycholinguistics by investigating the effect of interlingual homophones on word recognition and translation, examining processes involved in silent reading at the word level. Interlingual homophones are words that sound very similar in English and Brazilian Portuguese, but have different meanings, as is the case for ‘pie-pai’ and ‘value-velho’. In order to achieve the goals of this study, the general objective is divided into three specific ones: 1) To examine the processing cost of reading isolated Brazilian Portuguese-English interlingual homophones in relation to control words in a language decision task; 2) To investigate whether there is a repetition priming effect for isolated BP-En interlingual homophonic words in a translation task.; and 3) To examine whether there is a difference in the processing cost of reading isolated interlingual homophones from the L1 or the L2. The following hypotheses were proposed: H1 – Interlingual homophones between Brazilian Portuguese and English have a higher processing cost in relation to control words in a language decision task, reflected by longer reaction times; H2 – There are repetition priming effects for interlingual homophones in a subsequent translation task; H3 – Interlingual homophones from the participants’ L1 are recognized with greater ease in relation to L2 interlingual homophones. This research was carried out with a quantitative experimental methodology applied in real time through the online and free software PsyToolKit (Stoet, 2010, 2017), which allowed the data collection for two experiments: a language decision task, and subsequent multiple-choice task. The results of reaction times (RTs) and accuracy rates provided information about the cost of processing different types of stimuli. The present study is part of a larger project from the Laboratory of Phonetics and Multilingualism (LabFoM – UFC), which aims at conducting and disseminating experimental research in the fields of phonetics and language processing for bilinguals and multilinguals. The studies of Brysbaert et al. (1999, 2002), De Groot (2011), Diikstra et al. (1999, 2002, 2005, 2018), Haigh and Jared (2007), Toassi and Mota (2015), Van Assche, Brysbaert, and Duyck (2020), among other authors, provided theoretical support for the present research. The results of Experiment 1 showed that participants were significantly less accurate, but not less quick, to respond to interlingual homophones in comparison to matched controls only for the Portuguese language, partially supporting Hypothesis H1. Language decisions to English homophones were significantly faster than those to Portuguese homophones, opposing hypothesis H3. In Experiment 2, there were repetition priming effects only for control words in the accuracy data,

which opposes hypothesis H2. Moreover, homophone effects were not consistent across experiments, which suggested that these effects were modulated by task nature and also by prior exposure. The results provided further evidence supporting the language non-selective hypothesis predicted by recent models of bilingual lexical access and highlights that effects can be modulated by task requirements and language dominance.