RELAÇÕES DE DISTÂNCIA E DE COMPLEXIDADE ENTRE TRAÇOSDISTINTIVOS NA GENERALIZAÇÃO EM TERAPIA DEDESVIOS FONOLÓGICOS

by DUARTE, SABRINA HOHMANN

The general aim of this study was to verify the generalizations obtained by children withphonological deviations, submitted to the treatment proposed in the ABAB withdrawal andmultiple probe (TYLER & FIGURSKI, 1994), which takes the relations of distance andmarkedness among the distinctive features that were identified considering the targetsegment(s) used in therapy, as well as the segments absent in the child?s phonological system,as their parameters of analysis. These two types of relations between features have motivatedthe application, in this research, of two therapeutic models: (a) the MICT ? ?ModeloImplicacional de Complexidade de Traços?, proposed by Mota (1996), and (b) the MOTIDT ??Modelo Terapêutico Implicacional de Distância entre Traços?, which is being proposed inthis study. Both therapeutic models assume that the treatment, starting with a target segmentthat presents an internal structure with a complex feature configuration, makes its acquisitionpossible, and conducts to generalization, making the segments with simpler internal structureand feature configuration emerge. What differs fundamentally in the two therapeuticproposals is the criterion for the choice of the target segment of the therapy and theexpectation of generalization. In the subjects treated by MICT, the choice of the targetsegment(s) is derived from the relations of complexity established by the co-occurrencebetween the features and the ways the subjects increased complexity during the phonologicalacquisition with phonological deviations following the existing implicational laws betweenthe marked features, as described in the model. In the MOTIDT, the choice of the targetsegments is determined based on the distances among the features that integrate their structureand the structure of the missing segment(s) in the child?s phonological system. This distancewas determined following the feature geometry proposed by Clements & Hume (1995). Thesample of this work was composed of six subjects with phonological deviations, which weredivided into pairs, following the criterion of the degree of equivalency, considering theseverity of the deviation, constituting three pairs of subjects in the research. Initially, the datawere collected according to the proposal of Tyler & Figurski (1994), in which two sessionsare destined for the application of the Phonological Evaluation of the Child (YAVAS,MATZENAUER-HERNANDORENA & LAMPRECHT, 1991), and a session is reserved forcollecting spontaneous speech. After this, the data were transcribed and submitted to acontrasting analysis, based on Autosegmental Phonology (CLEMENTS & HUME, 1995), andan analysis based on the two therapeutic models used in the study: the MICT and theMOTIDT. The analysis of the subjects in this study have shown that the MOTIDT canconstitute an alternative therapeutic model for the clinical practice of treating phonologicaldeviation. The generalizations occurred in the phonological systems of the subjects treated bythis therapeutic model have reduced the treatment period once the subjects transferredknowledge of the target segments in a faster and more comprehensive form, when comparedto the patients treated under the MICT model. It is believed, however, that there are still morepoints to be explored and proven, with a greater number of subjects.