Efeito do significado sobre acurácia e tempo de resposta na comparação de segmentos

Abstract

Speech perception is conceived as a phase previous to lexical access. However, there is divergence among theories concerning this relationship. Some of them think of this sequence as being made of isolated processes, whereas others support that the processes are interactive and perform simultaneous or interactively. Nonetheless, there is not enough evidence to choose the ideal model. In this Masters thesis we tested if knowing the meaning of a word could influence the capacity of discriminating its sounds. The research was divided into three parts: a preliminary study to select items, the first data collection and the second data collection. The initial stage aimed at selecting the best set of pairs of monosyllabic pseudowords in order to make a task in which the capacity of discriminating consonants in the onset position of the syllable could be investigated. In the first and second data collections we investigated the capacity of comparing consonants in the onset position of the syllable before (pre-test) and after (post-test) a phase of training in which we assigned meaning to half the pseudowords of the task. The learning of meanings was verified between the pre-test and the post-test by using a specific task and only the subjects who scored 80% or more were included in the comparative analysis between the pre-test and the post-test. Both first and second data collections showed training effect in the comparison of segments, with a reduction in response time and an increase in accuracy (percentage of correct answers); however, there was no difference between the pseudowords which we assigned meanings to and those which we did not. Conclusion: Knowing the meaning of a word does not influence the capacity of discriminating its sounds.