Maternal speech to children with Down Syndrome: An update.

Abstract

Abstract Recent research on maternal speech to children with Down Syndrome (DS) generally corroborates studies conducted in the 1970s and 1980s. Specifically, the characteristics of language used by parents when speaking to children with mental retardation closely matches the characteristics of language addressed to typical children with the same language age. The question remains, however, why the language of children with DS develops more slowly and sometimes never reaches adult levels. This question is examined in reference to nativistic and learning theory frameworks. Keywords: language development, Down Syndrome, child-addressed speech, maternal speech, language environment, nativism, learning theory. Early Studies of Child-Addressed Speech Early studies of child-addressed speech examined the nature of maternal language input to typically developing (TD) language-normal children (Broen, 1972; Snow, 1972). Additionally, researchers analysed the verbal relationships between cognitively-normal adults and institutionalized adults with mental retardation (MR) (Siegel, 1963a,b,c, 1967; Siegel & Harkins, 1963; Spradlin & Rosenberg, 1964). In combination, these studies led researchers to question whether the language input to children with MR could be considered appropriate for their level of language development. Buium, Rynders, and Turnure (1974) assessed this question by comparing mother-child verbal interactions in a sample of TD children versus children with DS matched to the TD children for chronological age (CA=2 years). Data were collected in the laboratory while individual child-parent dyads engaged in collaborative play and in an instructional task. Results included statistically significant differences between the two samples of mothers regarding numerous aspects of the speech addressed to the children, and all differences suggested that the language addressed to TD children was more sophisticated than the language addressed to children with DS. The mothers of children with DS were observed to direct higher proportions of imperatives to their children than did the mothers of TD children. Buium et al. interpreted these results as demonstrating that the linguistic environment of languagelearning children with DS is different from that of TD children and therefore worthy of careful consideration in any attempt to understand the language delays of the former. Buium et al.'s interpretation was supported by independent research (e.g., Kogan, Wimberger, & Bobbitt, 1969; Marshall, Hegrenes, & Goldstein, 1973) showing that mothers of DS children are more directive and less likely to invite their child to take the lead in dyadic interaction than are the mothers of the CA-matched TD children. There was a shared belief in the specialized literature of the early- and mid-seventies that the differences in the language environment of TD children and children with MR had a negative influence on the language development of children with MR (e.g., Seitz, 1975; Mahoney, 1975; Mahoney & Seely, 1976; Mitchell, 1976; Levi & Zollinger, 1981). Such a belief was sometimes echoed without nuance as, for example, in Dolley (1974) who characterized adult and maternal speech to children with MR with respect to the restricted linguistic codes defined in Bernstein's early work on lower-social classes (Bernstein, 1961, 1964). The observation that mothers of children with MR tend to be more directive than mothers of CAmatched TD children (particularly in teaching situations) may not at all be due to a teaching style that is specific to mothers of children with MR. Instead, Rondal (1978a) argued that it may reflect the perceived necessity of these mothers to participate more actively in prompting the learning of their children with developmental delays. A major methodological problem in the early studies relates to the strategy of matching control group children by chronological age (CA). …