SPEECH·SOUND DISCRIMINATION AND ARTICULATION ERRORS IN CHILDREN AGED 8-12 by

Of the numerous variables which have been theoretically related tofunctional articulation disorders, speech sound discrimination (SSD) has received the most research and clinical attention. However, the majority of these investigations were done with children below age 7 while articulation therapy is generally consideredfor school-aged children. The present study investigated the possibility that in older children some misarticulations do not have this corresponding relationship. SSD was examined in 39 normal speaking children and 60 children with multiple misarticulation problems between 8 and 12 years of age. Significant differences were not found on SSD tasks between groups of normal children and those with articulation problems nor between groups of children with articulation problems categorized by type or number of phoneme error. Of the numerous variables which have been theoretically related to functional articulation disorders, speech sound discrimination (SSD) has received the most research and clinical attention. Nevertheless, the relationship remains controversial. When reviewing accumulated studies, Sommers & Kane (1974) concluded that: "the weight of experimental evidence indicates that children with functional misarticulations are likely to have inferior speech-sound-discrimination" (p. 114). Similarly, Winitz (1975) concluded that "the research literature has demonstrated that children with speech errors evince [sic] considerably more discrimination errors than normal speaking children" (p. 48). In contrast, Powers (1957) had previously found the experimental data to provide no evidence of Systematically inferior SSD in children with functional articulation defects and Rees (1973) more recently reported a similar, negative conclusion. McReynolds, Kohn and Williams (1975) obtained equivocal results from a distinctive feature analysis of misarticulating children's discrimination of their own production errors. Shelton, Johnston and Amdt (1977) and Shelton, Johnston, Rucello and Amdt (1978) were not successful in relating listening or discrimination tasks to articulation training and they, as well as W oolf & Pilberg (1971), have questioned the nature and underlying assumption of defective SSD in children with functional articulation disorders. The continuing controversy and conflicting findings argue against a simple association between SSD and articulatory proficiency. A number of investigators have proposed that Human Communication/Communication Hwnaine 6 (Spring) 1981 25 HUMAN COMMUNICATION, SPRING, 1981 SSD deficits may not be generalized but rather, related to specific phonemes (Prins, 1963; Aungst & Frick, 1964; Monnin & Huntington, 1974; Winitz, 1969; 1975), a possibility not explored in most previous studies. However, this hypothesis is supported by Locke's (l980b) recent finding in younger children that misproduction of the If-Ol contrast was highly associated with defective perception, whereas perceptual difficulty did not appear to underly productive failures involving the I w-11 contrast. The association between SSD and articulatory proficiency may also interact with age since the development of SSD decelerates markedly after age five, while articulatory development typically continues until the age of 8 (TempIin, 1957). Weiner (1967) concluded that a positive relationship between auditory discrimination and articulation is almost invariably found in studies of younger [italics added] children. Finally, Locke (1980b) noted that the substitution of If! for 101 was associated with misperception of this contrast much more frequently in younger than in older children. However, Sommers (1974) suggests that research is necessary to determine if "auditory deficiencies at early ages .. , continue into young adulthood" (p. 120). Deficits in SSD were early suspects as a basis for articulation problems (Travis & Rasmus, 1931). Training in SSD, commonly known as "ear training", was long ago (Van Riper, 1939) advocated as a therapy technique for functional articulation disorders. Despite the conflicting and equivocal research findings reviewed above, ear training has remained a common clinical practice (F1erning, 1971; Weber, 1971; Van Riper, 1972; Mower, 1980; Winitz, 1980). Indeed Winitz (1975) has stated that ear training alone may be sufficient for the correction of some articulation problems. Although originally associated with traditional approaches to articulation therapy, ear training has also been suggested in the context of more current approaches involving distinctive features (Winitz & Pressler, 1967), phonological rules (Ingram, 1976), and operant procedures (Mower, 1977). It appears that "many clinicians have faith in the common sense objective of ear training ... " (Perkins, 1977, p. 388), perhaps because some younger children are seen clinically to have difficulty discriminating between their misarticulated phonemes and the standard versions produced by the therapist. However, articulation therapy is most frequently carried out with school children aged 8 or older who are presumably past the age of phonological maturation (Templin, 1957; Sander, 1972). Thus, the supposition of SSD deficits and application of ear training exercises appears especially questionable for this group. No published account of articulation therapy differentiates between practices for younger and older children and standard therapeutic procedures appear to be applied without regard to age, phoneme or severity. Certainly if older children do not have difficulty with SSD current articulation therapy directed toward SSD or "ear training" would appear inappropriate. The present study was designed to examine the association between SSD and misarticulation of specific phonemes in older children with functional articulation disorders. To maximize the study's relevance to clinical practice, expe.rimental procedures included common features of articulation therapy. Thus, articulation was assessed on widely-used, commercially available tests and SSD was examined by requiring judgments of the similarity of paired stimuli, a paradigm familiar to most clinicians and incorporated into commonly used tests (Templin, 1957; Wepman, 1973). SUBJECTS Subjects were obtained by screening all children between the ages of 8 and 12 years in seven elementary schools. Subjects were selected from children found to meet the following criteria: 1) hearing thresholds of 20 dB or less (re: ANSI, 1969) in each ear at 500, 1000,