O efeito do treino de habilidades de consciência fonológica: sobre o aprendizado de leitura em um programa individualizado de ensino

Abstract

Alternatives literacy methods are encouraged by Education Ministry of Brazil. Aprendendo a Ler e a Escrever em Pequenos Passos is a computerized individual system to teach reading, developed for Brazilian children who exhibit protracted failure in school. It is based on stimulus equivalence paradigm (Sidman) and analyses of textual behavior from Skinner (i.e., by teaching individual words made of syllables, these minimal units can be recombinated, establishing generative reading). The program has four Units, made of sessions: matching to sample (MTS) and exclusion procedures designed to teach relationship between dictated and printed words, pre and post tests that evaluate the ability of reading trained and generalization words. Data shows the reading program effectiveness, mainly as regard to the trained words; even if it is observed return on generative reading, the results aren´t as good as wished. It is also observed that children usually have difficulty at the beginning of the learning; it implies that they need to repeat successively sessions until achievement of the settled criterion. According to scientific literature, the training of Phonological Awareness (PA) improves the reading skills. One study has demonstrated that students who were contiually unsuccessful at Aprendendo a Ler e a Escrever em Pequenos Passos, have recovered as soon as they have gone through a PA training. However, the impact of the PA training is not clear, once the students were subjected by both training - PA and reading - concomitantly. The purpose of the present study is to verify the real contribution of a PA training to this reading program; therefore, it was elaborated a training of PA abilities using MTS procedure, which was applied before Lendo e Escrevendo em Pequenos Passos. 13 students of the second and third grade from Elementary School participated in this research. Children were divided into two groups and accomplished different tasks: Experimental Condition performed the PA training before the reading program, while Control Condition executed a simple exercise of naming pictures. All students were tested three times (before the intervention, after the PA training or naming pictures task, and at the conclusion of first Unit of the reading program) through reading and PA´s evaluations. Tests grades, number of sessions to conclude Unit 1 and generative reading were compared. Results demonstrated similar performance between groups. Regarding the PA training, it seems that it has improved PA´s abilities, but it didn´t affect neither the learning of trained words nor generative reading. It is recommended extra researches with more participants, as well as some changes in the PA training (e.g., to consider some changes according to errors analyses done; inclusion of new tasks, as letters naming and letter-sound correspondence).