Ensino de leitura de sentenças para crianças em idade escolar
Abstract
Several teaching procedures based on the stimulus equivalence paradigm have been
developed and proved effective in improving reading performance. Expanding the research
on the subject, this study aimed to determine the effects of using a teaching reading sentences
procedure with this paradigm. The study used programmed sentences with multiple and
critical differences in order to check, as a secondary measure, possible patterns of responding
under the control of stimuli in this situation and the effects of these patterns on learning to
read. The sentences were composed of subject, verb and object. Ten children with seven to
nine years old, without clinical diagnosis, attending in the first primary education cycle,
participated. The study consisted of four phases and 12 steps. The sentences and the difficulty
of discrimination varied over the phases: simple words of the Portuguese language and
comparisons had multiple differences between them; simple words and critical differences;
complex words and multiple differences; and complex words and critical differences. The
steps contemplated pretest and posttest, teaching AB relations (match figures to spoken
sentences) and AC (match printed sentences to spoken sentences), testing of BD relations
(nomination of the figure), CD (nomination of the printed sentence), BC (match printed
sentences to the figures) and CB (match figures to the printed sentences), Reading Test of
recombinative generalization sentences, Reading Test of comparisons sentences and test of
learning maintenance after 30 days. Nine of ten participants had higher performance at
posttest than in the pretest. The children had more errors in the phases that had critical
differences. In the Reading Test of recombinative generalization sentences, two of ten
children had a high percentage of correct answers. The results suggest that the procedure
promoted the learning of sentences during the teaching; however, it was not sufficient for
recombinative generalization.