Response-class formation across French, Greek, and Turkish words in generalized vocal imitation by preschool children.
Item
-
Title
-
Response-class formation across French, Greek, and Turkish words in generalized vocal imitation by preschool children.
-
Identifier
-
AAI9807949
-
identifier
-
9807949
-
Creator
-
Kahya, Nursel.
-
Contributor
-
Adviser: Claire L. Poulson
-
Date
-
1997
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Psychology, Experimental | Psychology, Behavioral | Language, Linguistics
-
Abstract
-
Generalized imitation of words within French, Greek, and Turkish languages was investigated in three normally developing preschool children. During 20-minute experimental sessions vocal responses were modeled for the child by an adult. Models consisted of French, Greek, and Turkish words containing phonemes that do not exist in English. Six phonemes were selected to serve as discriminative stimuli to be modeled: two each from French, Greek, and Turkish languages. Then, for each of these phonemes, six two-syllable and six three-syllable words containing the phonemes in the initial position were identified. The dependent measures were the percentages of the targeted phonemes and the percentages of whole words per session that were matched by the child within three seconds of the model. Whole words were modeled, but reinforcement was contingent upon the child matching the targeted phonemes only. During each session 72 words were modeled. Of these, there were 54 words during training trials for which correct articulation of the phonemes was reinforced. There were 18 words during probe trials for which matching of the phonemes was not reinforced. Generalized imitation training was introduced in a multiple-baseline experimental design across the pairs of French, Greek, and Turkish phonemes. There were two experimental conditions: model-alone and model-and-reinforcement. During the model-alone condition the experimenter modeled the words with no differential consequences for matching the phonemes. During the model-and-reinforcement condition, matching of the phonemes produced reinforcement, and non-matching produced a correction procedure.;After the experimental question was answered, an extra training procedure was implemented for those phonemes for which the children had not been entirely successful. These extra training sessions occurred immediately prior to their regular experimental sessions, which remained unchanged. During the extra training procedure only the targeted phonemes were modeled and reinforcement was contingent upon correct phoneme productions.;There were three primary findings. First, reinforcement of phoneme matching produced primary generalization of the targeted phonemes from the trained word context to the non-trained word context. Second, reinforcement of phoneme matching produced generalized imitation of words during training trials. Third, reinforcement of phoneme matching during training trials produced generalized imitation of words during probe trials. Implications concerning primary generalization and the formation of response classes within the class of vocal generalized imitation are discussed.
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
