Repetition of Comprehensible Sentences by Children with Deviant Speech

Abstract

Recently the imitation of sentence strings has been found useful in probing the linguistic abilities of language disordered and normal children. The impetus for this method was provided by Chomsky (1964), who regarded repetition performance as one indirect way of tapping linguistic competence for the syntactic, semantic and phonological rules of grammar. The technique he suggested was to have the child repeat "sentences and nonsentences, phonologically possible sequences and phonologically impossible ones" for the analysis and interpretation of error patterns across levels of the grammar. Correct placement of elements would suggest acquired competence, while misplacement would indicate lapses in performance, or nonacquisition of linguistic competence. In this manner, Chomsky speculated, it might be possible to determine whether fully developed conceptions of sentence structure exist under the telegraphic speech patterns so commonly observed in young children. Subsequent repetition studies on the whole have confirmed both the utility of the repetition method and some of Chomsky's predictions. Fraser et al. (1963), using an imitation-comprehension task, were able to differentiate a speech system from the grammatical system. With the acquisition of linguistic competence, speech repetition comes under the control of the linguistic system, so that the structure of imitations reflects the latter rather than the former. If the normal child is given a syntactic string within the limits of his short-term memory performance, accuracy of repetition will be determined by semantic (Fraser et al., 1963) syntactic (Menyuk, 1969) and phonological (Whitacre et al., 1970) complexity. In short, the grammatical system "filters" the input data (Menyuk, 1969). In the case of children with deviant speech, the amount of filtering is greater than for the normal child. They seem to be particularly affected by transformational complexity and limitations on short-term memory performance (Jackson and Sommers, to appear;Menyuk, 1969 ;Whitacre et al., 1970). Omission errors predominate, although essential semantic properties are left intact (Menyuk, 1969).The purpose of this study was to examine the sentence repetitions of a group of children with normal decoding abilities, but with severe deficits in speech-motor performance as reflected in their largely unintelligible speech as perceived by clinicians and others. Their grammatical expression was viewed as an extreme form of telegraphic speech (McNeill, 1970). The repetition data were available from a companion study in which these children were found to understand sentence commands as spoken by peers with intelligible speech, but not as spoken by themselves (Panagos and King, to appear). Similar results had been found in a pilot study (Tarulli and Panagos, 1970). Thus the particular aim here was to grammatically analyze imitated sentences for which normal speech comprehension had been established. A secondary goal was to relate the grammatical data to symptoms of verbal apraxia typically rep...

Contextual and Linguistic Correlates of Intelligibility in Children With Developmental Phonological Disorders

“…The interpretation that decreased intelligibility for words associated with phonetic complexity, syllabic complexity, and certain grammatical forms is mediated by increased speaker errors is supported, respectively, by many studies that have documented: (a) increased deletion errors associated with consonant clusters (Dukes & Panagos, 1973;Hodson & Paden, 1981;Oiler, Jensen, & Lafayette, 1978;Shriberg et al, 1986), (b) increased and more atypical articulation errors in polysyllabic words (Klein & Spector, 1985;Shriberg et al, 1986), (c) increased articulation errors in unstressed contexts (e.g., function words; Campbell & Shriberg, 1982), and (d) disproportionally better articulation of nouns by children at the early-word stage (Camarata & Leonard, 1986). Comparable findings have been reported for subgroups of speakers with motor speech disorders, with increases in articulatory errors related to variables such as word length, grammatical class, and sentence position (Darley, 1982).…”