Ensino de relações de equivalência com sentenças de cinco termos e produção oral em uma criança com implante coclear1

Abstract

Previous studies showed the effect of strengthening of relations between printed sentences, actions pictures and dictated sentences in oral naming of pictures with point-To-point correspondence. The sentences had three terms. The present study evaluated if strengthening relations between dictated sentences, pictures of actions and printed sentences promove the accuracy in oral naming of action pictures for a child cochlear implant' user and rudimentary reading skills. Different from previous studies, two sets of sentences with five terms each were used; the stimuli were organized on matrix with overlapping of third term (objetc); the first (subject) and second (verb) terms were permuted same position in various sentences. The experimental steps were conducted: pretests of reading and of naming involving sentences; teaching of conditional relations between dictated sentences and action figures (by matching to sample, MTS) and construction of printed sentences by dictated sentences (via constructed-response matching to sample, CRMTS); post-Tests; teaching revision with echoic training, if necessary. In the pretests of set 1, the participant obtained 54.55% correct responses in reading and none in naming; in post-Test, she got 76.36% correct responses in reading sentences and zero percent in naming; only after teaching revision with echoic training, she got 80% correct responses in naming. The results of set 2, there was less exposition to teaching relations and without echoic training. The results replicate findings with 3-Terms and extend for 5-Terms. The effects of echoic and of strengthening relations in equivalence-based network of reading in speech accuracy of sentences in naming were noted in children with cochlear implant and basic skills of reading. The inclusion of articles can made difficult, initially, the control establishment for stimuli sentence. Future research should control the variables effects and verifcate the generality with more participants.