A tradução de sinais em Libras para o português sob a ótica da psicolinguística

Abstract

This study investigated the cognitive processes involved in translation between the Brazilian Sign Language (Libras) and Brazilian Portuguese language pair, from a psycholinguistic perspective. Translation between Libras and Portuguese involves specific competencies, given that they are languages of different modalities. The general objective of this study was to investigate lexical access in Libras-Portuguese translation by Libras interpreters through response time and accuracy measures. Using a quantitative experimental approach, an experiment was constructed with a corpus of 40 Libras signs, characterized as real and unreal. A hand configuration was selected as a parameter to be subjected to variation and form the unreal signs. Substantive signs without semantic relation were selected. For sign validation, the Capovila trilingual dictionary (2002) and the INES online dictionary (n.d.) were used. The research was conducted online using the PsyToolkit tool (STOET, 2010, 2017). The 70 participants performed two tasks: one lexical decision task and another multiple-choice translation task, where they viewed three translation options and had to choose the correct item. After completing the tasks, the collected data on response time (in milliseconds) and accuracy were processed using the R-Studio tool. In the lexical decision task, the analysis showed significant differences between the experimental conditions (real and unreal) regarding response time and accuracy, indicating that understanding Libras (Brazilian Sign Language) involves complex processing, and that cognitive effort increases when the sign does not belong to a known linguistic system, thus strengthening Libras' status as a natural language with all its linguistic parameters. The analysis of the multiple-choice translation task showed a significant difference between the conditions only for accuracy, but not for response time. This result partially confirms the initial hypothesis that participants would require more response time when translating real signs when these signs were seen in the unreal condition during the first task. It is worth highlighting the relevance of this research, given the scarcity of studies on cognitive processing in translation between Libras (Brazilian Sign Language) and Portuguese, especially from a psycholinguistic perspective. Understanding these processes is fundamental to improving the training of translators and interpreters, and to contributing to theoretical advancements in Translation Studies and Psycholinguistics applied to sign languages.