Journal of Speech and Hearing Research, Volume 30, 503-521, December 1987 INDIVIDUAL DIFFERENCES IN PHONOLOGICAL DEVELOPMENT: AGES ONE AND THREE YEARS MARILYN MAY VIHMAN MEL GREENLEE Stanford University This paper reports the results of a study of the persistence of individual differences in the phonological development of 10 normally developing children observed at age 1 year and again at age 3 years. Data were based on 1/2-hr audio and video recordings of weekly spontaneous mother-child interaction sessions in the home between 9 and 17 months and at 36 months. In addition, phonological and cognitive probes were administered at age 3. At age 1 the children were compared at four times selected on the basis of the number of different word types used in a session. Preferences for particular phonological categories (fricatives, liquids, final consonants) were found not to correspond to relative mastery of those categories at age 3. Based on both babble and words, high use of vocalizations containing true consonants was found to be predictive of greater phonological adVance at age 3, Phonological errors of two kinds were distinguished for age 3: those resulting from difflculty with specific segments and those more typical of younger children, involving the rearrangement, assimilation, or deletion of segments or syllables (prosodic errors). The children differed in intelligibility and in specific segment substitutions and cluster reductions. They also differed in the proportion of prosodic errors made and in consistency in segmental errors. Lastly, aspects of cognitive or learning style as expressed in phonological organization were found to be recognizable at both age 1 and age 3. The possibility that children may follow different paths or strategies in language acquisition was noticed as early as the 1960s, when the prevailing emphasis in the field was on universal aspects of development. One of the 3 children in the classic Harvard study of the acquisition of syntax, Eve, advanced more rapidly than the other 2 in terms of mean length of utterance, but she also combined more content words and fewer grammatical words and inflections at a given point in syntactic advance than did Sarah, who reached the same syntactic milestone at a much later age (Cazden, 1968, 1972). Since then there have been numerous accounts of variation across subjects in the acquisition of Syntax (cf. Bloom, Lightbown, & Hood, 1975; Horgan, 1980) and the lexicon (cf. Nelson, 1973). Rare attempts to integrate individual approaches to more than one domain include Bretherton, McNew, Snyder, and Bates (1983) and Vihman and Carpenter (1984). Relatively fewer studies have been devoted to individual ,variation in the acquisition of phonology, and those few have been almost exclusively focused on the earliest stages, that is, on the 1 year-old (cf. Ferguson & Farwell, 1975; Leonard, Newhoff, & Mesalam, 1980; Shibamoto & Olmsted, 1978; Stoel-Gammon & Cooper, 1984). Recent studies of the phonological systems of 1-year- old children have produced general agreement that indi- vidual differences are a significant factor at that stage of development, although the particular strategies and pos- sible sources of the differences have varied. On the basis of their analysis of initial consonant use in the first 50 words of 3 children, Ferguson and Farwell (1975) re- ported two major findings: the exploitation of different "favorite sounds" by different children, and the use by some children of well-defined lexical selection on pho- nological grounds, leading to rapid lexical expansion at the expense of phonological differentiation. Shibamoto and Olmsted (1978) expanded Ferguson and Farwell's analysis to medial and final position (for 4 subjects) and confirmed the existence of lexical selection, especially on the basis of the initial consonant. They also reported different individual strategies ("fronting", "reduplica- tion", "centralization"). Leonard et al. (1980) further expanded the data base, analyzing consonant use in the spontaneous word produc- tions of 13 subjects, including a pair of identical twins and a hearing child of deaf parents. They too found differ- ences among the children, but emphasized that these differences fell within a relatively narrow range. On the basis of their study of the phonology of the twins, who differed as much as any other pair of children despite exposure to much the same input, Leonard et al. con- cluded that the linguistic environment cannot be the chief source of individual differences in phonology. Fi- nally, the lexical factor highlighted by Ferguson and Farwell (1975) was also found to be important in this broader-based investigation. Most recently, Stoel-Gammon and Cooper (1984) reported data from the babbling, in- vented words, and first 50 adult-based words of 3 subjects. The children were found to vary most along three parame- ters: (a) rate of phonological advance, (b) use of a lexical selection strategy, and (c) relative accuracy or variability in phonological production. Grunwel] (1981) presented a survey of the current state of knowledge about normal phonological development in the form of a profile covering the period from first word- use (9 to 18 months) to age 4 (see also Grunwell, 1982). She included a useful chronology of phonological pro- cesses, indicating the age by which each of the simplify- ing processes may be expected to have ceased to apply (see Figure 1). Grunwell noted that a wide range of individual variation can be documented for the earlier stages of development. For the later stages, however, large scale studies reporting group results have been the rule, resulting in a dearth of information regarding indi- © 1987, American Speech-Language-Hearing Association 503 0022-4685/87/3004-0503501.00/0 Downloaded From: http://jslhr.pubs.asha.org/ by a Univ of York-England User on 07/28/2015 Terms of Use: http://pubs.asha.org/ss/rights_and_permissions.aspx