A relação entre a compreensão e os aspectos prosódicos na leitura em voz alta de falantes do PE e do PB

Abstract

The main goal of this paper is to analyze the reading aloud of subjects of different ages/school grades, speakers of European Portuguese from the city of Lisbon and speakers of Brazilian Portuguese from the city of Belo Horizonte and its relation to comprehension of the material being read.Thirty subjects were selected from each variation, grouped by age, 10 with 11 years of age, 10 with 15 years of age and 10 with 20 years of age. Each subject read aloud two texts twice. Between the first and the second reading, the subjects answered, in a computer, a comprehension test. To categorize the subjects as fluent and not fluent, the prosody indicators measured and analyzed in each text were the duration of the reading and the speech rate.From each text some sentences were selected and speech rate and duration, articulation rate and duration, pauses length and percentage, the number of intonational syntagms were analyzed. All these data were compared to two pre-selected control subjects, one from each variation, considered fluent readers. These comparisons showed that the among the PE subjects, only the 15 and 20 years old can be considered fluent. The 11 years old showed a significant difference from the others. Among PB subjects, the results were quite different. When compared to the control subject, the only significant difference found was in pauses length and percentage, this difference was also noted when comparing subjects from each age group. It is important to notice the difference in the use of pauses between the PE and PB: in PE the pauses seem to show the subjects fluency and no fluency, as for PB, pauses limit the intonational syntagms limited, more frequently, in PE, by the high tone border H%.The comprehension data showed that one group could present very different results for texts of different genres. The more complex text allowed us to see an inversely proportional relation between age/school grade and the number of errors in the comprehension test. The older subjects presented a smaller number of errors. In the beginning of the research, we were expecting that reading aloud fluency was a consequence of good comprehension. However, our results showed that this relation should be rethought. The 20 years old subjects, in both PE and PB, showed fluency while reading aloud, but they also showed a high number of errors in the comprehension test related to the simpler text, therefore we concluded that fluency and comprehension arent always directly related.