Language and Communication Characteristics of Children with Down Syndrome

New Perspectives on Down Syndrome Jon F. Miller, Chapter 9, 233-62

Excerpted with the permission of David Tolleson © 1987 by the National Down Syndrome Congress. All rights reserved. 1370 Center Drive, Suite 102 Atlanta, GA 30338 (800) 232-6372

Children with Down syndrome are particularly at risk for language learning problems for reasons beyond the associated cognitive deficits. First, there is an increased frequency of middle ear infection, which is frequently associated with delayed language acquisition in normal children (Brandes & Elsinger, 1981; Downs, 1980). Frequent middle ear infection can result in hearing loss, which is always associated with language learning problems. Second, the deficits in motor coordination associated with Down syndrome may adversely affect the synchrony of motor movements required of the speech production system, including respiration, phonation, and articulation of the palate, tongue, lips, and jaw (Bless, Swift, & Rosen, 1985). Third, cognitive deficits specific to Down syndrome may result in language learning problems beyond those commonly associated with mental retardation (Miller, Chapman, & MacKenzie, 1981). And fourth, there can be decreased expectations for performance of mentally retarded individuals, which frequently result in learned incompetence or lack of appropriate experience (Coggins & Stoel-Gammon, 1982). Any one of these factors can result in deficits in language acquisition; taken together, they represent a formidable puzzle to unravel in order to understand the forces affecting language growth in this population.

The focus of this chapter is language learning in individuals with Down syndrome, with a consideration of all of the factors affecting the acquisition and use of language as a symbolic system. Prior to the discussion of the studies of children with Down syndrome, the general themes in the research on the language characteristics associated with mental retardation are reviewed. These themes include investigations of cognitive characteristics and the extent to which they are necessary for language learning, and investigations of the nature of the language acquired by individuals with mental retardation (i.e., is it delayed or deviant relative to normal acquisition?). Investigations of the speech and language performance of children with mental retardation have focused, in general, on the major determinant factors for acquiring a language system, including cognitive constructs, environmental facilitation, and physiological limitations, both perceptual and motor.

LANGUAGE ACQUISITION AND MENTAL RETARDATION

There can be little doubt that children with mental retardation do not demonstrate the same language behaviors as normal children of the same chronological age. However, despite a variety of methods used to study the various etiologies resulting in mental retardation, the results of most studies indicate that children with mental retardation use normal linguistic forms and do not produce bizarre language patterns, such as unique word combinations, invented word meanings, or novel discourse characteristics (Duchan & Erickson, 1976; Graham & Graham, 1971; Kamhi & Johnston, 1982; Lackner, 1968; Miller et al., 1981; Naremore & Dever, 1975; Newfield & Schlanger, 1968; Ryan, 1975; Semmel, Barritt, Bennett, & Perfetti, 1967; Yoder & Miller, 1972). Mental retardation is a behavioral classification resulting from a variety of etiologies, including: metabolic and genetic syndromes; disease processes; trauma, both pre- and postnatally; and cultural and familial influences (Miller et al., 1981). It is impressive that despite a variety of brain syndromes, children with mental retardation learn the standard form of their native language.

Attempts to characterize the language performance of persons with mental retardation have evolved over time, resulting in several competing points of view over the past 25 years. Initially, a major issue was whether the language of these persons was quantitatively or qualitatively different (i.e., delayed or deviant relative to nonverbal cognitive skills). This controversy was derived in large part from different ideas about the relation between language and cognitive skills. The view of quantitative differences is consistent with the idea that developments in language are directly contingent on developments in cognition. Such a view predicts that language skills are always consistent with cognitive development or mental age, and never advanced or delayed relative to nonverbal mental age development (Graham & Graham, 1971; Lackner, 1968; Lenneberg, Nichols, & Rosenberger, 1964). Further, quantitative differences argue for similar cognitive structures, with learning rate the primary problem, which results in language learning being characterized as "slow motion" normal development. The qualitative or deviant development view of the language of children with mental retardation argues that language learning is delayed relative to the performance level predicted by mental age (Newfield & Schlanger, 1968; Semmel et al., 1967). Delays in language development relative to mental age are interpreted as deviant language performance, compared to the synchrony of development of chronological age, mental age, and language in normal children. Several studies have supported the deviant view. However, the majority of research has supported the position that language development in children with mental retardation follows the same course and sequence as that of normal children but progresses at a slower rate (i.e., it is quantitatively different) (Miller & Yoder, 1974; Ryan, 1975; Yoder & Miller, 1972). In general, the language performance of children with mental retardation has been found to be similar in form and content to that of normal children matched for mental age or linguistic stage of development.

Both the delay and deficit views recognize that cognitive development is essential for language development. The delay position implies that development in cognition is sufficient for language to develop, and the deficit view recognizes that additional child characteristics are necessary for language growth. Both of these views, however, oversimplify the complexities of the language itself and the acquisition process, viewing it as unidimensional and static over time. Neither considers the role of social factors, the impact of language on cognition, or the influence of specific linguistic capabilities on language development, which are prominent ideas in current theories of language development (Bates & MacWhinney, 1979; Slobin, 1985; Wells, 1980). Recent surveys of the relation between language and cognition (Cromer, 1981; Finch-Williams, 1984; Leonard, 1978; Miller, Chapman, Branston, & Reichle, 1980) have suggested a correlational relation, in which language development proceeds generally at the rate of cognitive development, but depending on environmental events, may be slightly behind or ahead of cognitive skills. Such a view recognizes the central role the child's environment must play in language acquisition, which includes providing both quality and frequency of language and nonverbal experiences leading directly to increasing knowledge of the world.

Studies of children with mental retardation have supported this correlational view of the language-cognition relationship, documenting differences in language performance not accounted for by cognitive status. Two studies have investigated the profiles of language comprehension and production in relation to nonverbal cognitive development in children with moderate and severe mental retardation (Miller, Chapman, & Bedrosian, 1978; Miller et al., 1981). Significant asynchronies were found in the profiles of these children, who were functioning developmentally between 7 months and 7 years. The studies found three profiles of language skills: 25% of the subjects had delays in productive language relative to comprehension and cognitive skills, 25% of the subjects had delays in both comprehension and production of language relative to cognitive skills, and 50% had comprehension and production skills commensurate with cognitive skills. One-half of the subjects studied (i.e., the first two groups) did not match the generalization of retarded performance (i.e., "slow motion" normal performance). The delays could not be attributed to hearing or speech-motor deficits. They were evident at several levels of cognitive skill, from Piaget's sensorimotor stage V through late preoperational development (Piaget & Inhelder, 1969). The subjects at each cognitive stage were of various chronological ages. Perhaps language skills fail to develop at the same pace as other cognitive skills in some children as they get older either due to etiology or other intervening factors. The variation in development found in individual children across specific linguistic features (vocabulary, syntax, and semantics in both comprehension and production) may be due to social or environmental differences.

In another study, Kamhi and Johnston (1982) found frequency differences among linguistic characteristics they attributed to social and motivational behaviors rather than deficits in cognitive or linguistic abilities. This study is important in that it proposes different underlying processes for different language characteristics. Frequency of use of specific linguistic forms is attributed to social aspects of the conversational context or to motivation. Differences in frequency of use would not be interpreted as reflecting differences in linguistic or cognitive knowledge. Taken together, these studies suggest that the social interaction, including language input and maternal responsiveness, may explain the individual variation in language learning in children with mental retardation.

To summarize, previous research on children with mental retardation has resulted in several very general conclusions about their communication. The language performance of such children has been described as following the same developmental pattern as normal children when matched for mental age or linguistic stage. Despite differences in etiology, the majority of research has documented similar performance across linguistic domains. While language and communication skills are clearly delayed relative to chronological age, many of the differences disappear when children with mental retardation are compared to children of similar mental abilities. Other studies (Kahmi & Johnston, 1982; Miller et al., 1981) have documented differences in language not attributable to cognitive deficits. In investigating language and communication skills in children with Down syndrome, one must evaluate language skills relative to their underlying processes, cognitive as well as social, recognizing the complexity of the language system as it is used in speaking and listening.

LANGUAGE AND DOWN SYNDROME

The research literature dealing with the language behavior of children with Down syndrome has reviewed many of the same issues associated with mental retardation in general, including the components of language behavior and the internal and external (cognitive and environmental) forces associated with language development. The goal of this chapter is to document what is known about language acquisition in Down syndrome so that improved methods for facilitating language growth and change can be developed.

What specific areas of expertise must be developed in order to learn a language? Consider that the child must acquire the sound system, vocabulary, and grammar to convey an unlimited set of meanings derived from his or her knowledge of the world. Learning and using a language system requires the development of both comprehension and production skills to allow its use in socially appropriate ways-to give and get information in conversation, tell stories, argue, persuade, and accomplish all of the other functions for which language is used. Normal children acquire their language over the course of childhood, with more advanced communicative skills added in the early teenage years (deVilliers & deVilliers, 1985). Children demonstrate linguistic performance at the onset of the 2nd year, comprehending and producing single word utterances. Development proceeds rapidly through age 5, when the basics of syntax for simple sentences have been mastered. The ability to convey a range of information-including marking past and definite future events, simple spatial relationships, and primitive causal constructs-has also developed. The essential elements of conversation skills have also been mastered. Vocabulary size is estimated to be 23,700 different words at age 7 (Smith, 1941), indicating an accelerating vocabulary acquisition rate of 7-10 new words a day. Any characterization of retarded language development must view the individual features of the language system (i.e., phonology, syntax, semantics, and pragmatics) as they develop independently and in synchrony with each other in speaking and listening.

As mentioned previously, there are several characteristics associated with Down syndrome that put these children at particular risk for language deficits beyond their cognitive deficits. They have increased incidence of otitis media resulting in hearing loss. They may have speech-motor control problems resulting in deficits in speech intelligibility. There may be specific cognitive deficits associated with language learning, and differences may exist in expectation for performance. All of these characteristics will affect language interaction patterns, stimulation, and responsivity. The requirements for language development include intact perceptual, cognitive, and speech-motor systems, and a stimulating and responsive linguistic environment (Miller, 1981). Children with Down syndrome are at risk for language problems resulting from potential deficits in any one of these essential areas. The majority of work viewed in this chapter compares the performance of children with Down syndrome to normal children in order to address the following question: Are children with Down syndrome acquiring language at the same rate and in the same sequence as normal children? In the majority of studies, normal children and children with Down syndrome have been matched on chronological age, mental age, and, recently, language level in an effort to uncover any differences in language performance between the two groups. Baumeister (1967, 1984) pointed out the dangers inherent in matching studies, particularly in inferring underlying processes from similar to different performance characteristics between normal children and children with mental retardation. While the comparative studies are, necessarily, reviewed in detail, the parallel theme of this chapter is the emerging language competence of children with Down syndrome.

DO CHILDREN WITH DOWN SYNDROME HAVE A DISTINCTIVE PATTERN OF LANGUAGE DEVELOPMENT?

The literature on productive language development of children with Down syndrome reveals a pattern of increasing linguistic deficit with advancing chronological age (CA). A summary of the results of 16 studies can be found in Table 1. As chronological age advances, deficits in language development below mental age (MA) expectations are more likely to be found. Table 1 also includes the developmental age or linguistic stage of the persons studied, when available. Together, these studies investigated 226 individuals with Down syndrome, ranging in chronological age from 6 months to 49 years.

Syntax and Semantics

Several studies have reported that the language skills of children with Down syndrome fail to keep pace with their increasing cognitive abilities (Harris, 1983; Rogers, 1975; Wiegel-Crump, 1981). Wiegel-Crump (1981) studied the growth of syntax in 80 children with Down syndrome, with CAs of 6-12 years, and nonverbal MAs of 2-6 years. While she found no difference between home-reared and institutionalized subjects, she found significant MA effects at each level (2, 3, 4, 5, and 6 years). Compared to the norms on the Developmental Sentence Scoring task (Lee, 1974), children with Down syndrome with MAs of 2 years performed at the 50th-60th percentile and subjects with MAs of 3-6 years performed at the 10th-15th percentile. Little change was noted in the language skills of the 3-6 year MA group. Their language was characterized as stereotypic, repetitive, and inflexible. Rogers (1975) reported that of 17 subjects tested on the Raynell Test of Language Performance, all 17 had scores below those of MA comparisons for language structure, 13 of 17 subjects had lower scores on vocabulary, and 11 of 17 subjects had lower scores on language content. Six of 17 subjects had good speech intelligibility, and a significant relation was found between speech intelligibility and language structure scores. Surprisingly, Rogers found the comprehension scores of the older children with Down syndrome (10-16 years) to be significantly worse than their production scores. All of the children with Down syndrome showed significant language deficits relative to their nonverbal mental abilities. These two studies, which investigated almost 100 subjects with Down syndrome, using cross-sectional research designs, document a pattern of increasing linguistic deficit with age.

The strategy of matching normal children and children with Down syndrome on general language ability is used when relations among language characteristics are being investigated and when differences in language development would confound the results. Usually investigators use mean length of utterance (MLU), which is a calculation of the child's average spoken utterance length in words or morphemes. This measure is interpreted as an index of general syntactic development (Brown, 1973; Miller, 1981), and therefore allows comparisons of specific syntactic devices or semantic and discourse features. Several studies of younger children with Down syndrome have documented similarities in language and communication abilities when directly compared to normal subjects matched on then- stage of language development by MLU. (Note the developmental level of these children in Table 1.)

Harris (1983) studied 10 children with Down syndrome and 10 normal children to determine the validity of MLU as a linguistic matching technique. His subjects with Down syndrome ranged in CA from 2.6 to 6.9. No MA data were reported. He found that the language of his younger subjects with Down syndrome closely resembled the language of normal children of the same MLU stage. However, the older subjects used different means to expand their utterances, suggesting different language learning strategies that are evident only as both children with Down syndrome and normal children become older and acquire more advanced language skills.

Coggins (1979) found that children with Down syndrome and normal children expressed the same range of meanings, as coded by semantic relation categories in extensive free speech samples. The four subjects with Down syndrome were in linguistic stage one, MLU = 1.0-2.0, the earliest stage of language development. Normal children pass through this stage between 12 and 24 months of age. Rondal (1978b) found no differences between the language performance of his 21 subjects with Down syndrome as compared to the language of normal children matched for MLU. As MLU increased, language complexity increased, expressing the same range of meanings. The children with Down syndrome showed more advanced vocabulary development than the normal children at the same MLU stage. Owens and MacDonald (1982) studied the expressive communication ability of six subjects with Down syndrome and six normal subjects at linguistic stages I and III, MLU = 1.0-2.0 and 2.5-3.0, respectively. No differences in the range or frequency of communicative behaviors expressed verbally were found. The children with Down syndrome and the normal children used language equally well across a range of communicative situations (e.g., requesting, commenting, answering, naming). Coggins, Carpenter, and Owings (1983) found similar results investigating four subjects with Down syndrome and four normal subjects all at linguistic stage I, in sensorimotor stage VI, and having MAs of 18-24 months. They coded communicative behavior which was expressed either gesturally or verbally. They found that both the range and proportional use of communication behaviors were the same for the subjects with Down syndrome and the normal subjects.

The often reported preference for gestural expression over verbal expression among children with Down syndrome by educators, parents, and speech-language pathologists has been examined by several investigators. Using a well-documented coding scheme, Coggins and Carpenter (1981) reported similar use of vocal and gestural means of expression. Greenwald and Leonard (1979), however, found that children with Down syndrome, as compared to normal children of similar language and cognitive stages, used more gestures than verbal means in expressing imperatives and declaratives. Coggins et al. (1983) studied the free speech of their subjects while playing with their mothers. Greenwald and Leonard (1979) used an elicitation technique developed by Snyder (1978). Differences in performance may be due to differences in the communicative demands of the two situations, one coding naturally occurring events, the other setting up contingencies to evoke a communicative response. One could speculate that when communication is demanded, the children with Down syndrome use gestures over words, but that they can express the same range of communicative events verbally as their normal counterparts when they are not under communicative pressure.

Phonology

In an investigation of 10 children with Down syndrome, premeaningful vocalizations for the first 15 months of life were reported by Smith and Oiler (1981). They found substantial similarities between children with Down syndrome and normal children. Both groups began to produce canonical, reduplicated babbling at 8-8½ months of age. The development of vowels and consonants showed similar patterns for the first 15 months of life. Stoel-Gammon (1980) reported on the articulation problems of four children with Down syndrome in Brown's (1973) early and late stage I of linguistic development. The articulatory patterns were reported in relation to their overall language performance, as well as their articulation errors in terms of phonological processes. The results indicate that while subjects are able to produce nearly all of the consonant phonemes, correct production in conversational speech was usually limited to a particular position in the word; the position varied across subjects. The phonological processes that account for the speech sound errors are similar to those reported for young normal children. The phonological abilities of the four children with Down syndrome are comparable to or better than their general language ability, as measured by their mean length of utterance.

The findings of Smith and Oiler (1981) and Stoel-Gammon (1980) confirmed that children with Down syndrome exhibit knowledge of the phonological rules for speech-sound combination commensurate with their general language ability. Phonological rule knowledge may be appropriate for those children with Down syndrome who have fairly intelligible speech. Speech intelligibility remains, however, a consistent problem among individuals with Down syndrome. Intelligibility problems can be directly attributed to the oral and facial anomalies associated with the condition, affecting the jaws, tongue, teeth, and gingival and mucosal structures (Sanger, 1975) and the speech-motor control of these structures (Bless et al., 1985). The tongue is frequently reported to be too large for an undersized oral cavity and protruding from the mouth. Motor control of all oral structures is negatively affected by the general hypotonia associated with Down syndrome. In addition, there is evidence for a general motor deficit in Down syndrome that particularly affects coordination and timing of motor movements. The range of speech intelligibility in the syndrome is, therefore, attributable to both structural and motor control deficits.

Intelligibility of speech intrudes on all of the other aspects of productive language development discussed. Given the range of intelligible speech, one wonders how subjects were selected for a study of productive language. Informal surveys of researchers frequently result in comments about the high frequency of unintelligible speech. Paul Berry (personal communication, 1984), for example, who is conducting a longitudinal study of 60 children with Down syndrome and their families in Queensland, Australia, reported that the majority of his 5- and 6-year-olds were unintelligible. Elicitation tasks were required to study productive language because, without knowing in advance what was being said, their speech was not interpretable. Better reporting of subject-selecting criteria must be made if the results are to generalize to the full population of children with Down syndrome, rather than only to those who have mostly intelligible speech.

Discourse

Discourse is defined in this chapter as any analysis of language units larger than an utterance, particularly the contingent relations between utterances in conversation, narration, or other pragmatic speaking conditions. In an investigation of children's strategies to clarify utterances when they are not understood by a speaking partner, Coggins and Stoel-Gammon (1982) found that all four of their subjects with Down syndrome were sensitive to their speaking partner's request for clarification. This was evidenced by either a repetition or a revision of the utterance. The results document that by linguistic stage I, children with Down syndrome have mastered this major aspect of human communication, as would be expected of normal subjects at the same level of linguistic development.

Leuder, Fraser, and Jeeves (1981) and Price-Williams and Sabsay (1979) have documented the advance of communicative competence over linguistic skills in adult subjects with Down syndrome. These studies investigated conversational skills, discovering that subjects used a variety of strategies for gaining attention and clarifying utterances when they were not understood. The subjects tended to use discourse strategies to establish topic, and relied on posture and gestures to indicate appropriate social distance with strangers. In general, subjects with Down syndrome used nonverbal interaction strategies over verbal means of communication. Both Price-Williams and Sabsay (1979) and Leuder et al. (1981) concluded that adults with Down syndrome lack the linguistic means for expression of their experience and knowledge of the world, but that their communicative competence is intact to the extent that their severe linguistic deficits permit.

Another study of the discourse skills of three subjects with Down syndrome documented advanced communicative skills when language skills more closely correspond to mental abilities (Nisbet, Zanella, & Miller, 1984). The three subjects in this study were 12 and 15 years of age and, educationally, were mainstreamed into a normal high school. Free speech samples were recorded in conversation with peers and with a normal student of the same chronological age. All of the subjects were rank-ordered according to their social and communicative competence. Analysis revealed that all of them demonstrated more advanced linguistic competence as they assumed more competent speaking roles, by adjusting their language ability to take charge of the conversation, asking more questions, introducing more topics, and using longer utterances. This study complicates the view of increasing linguistic deficit with age, clearly contradicting the trend noted in the majority of studies reviewed. Where language skills have developed in children with Down syndrome, overall communicative competence appears to be finely tuned.

Studies of chronologically younger children with Down syndrome have documented similar development of language structure, semantic expression, and discourse features compared to normal children matched for MA and general language level (Coggins, 1979; Coggins et al, 1983; Coggins & Stoel-Gammon, 1982; Greenwald & Leonard, 1979; Harris, 1983; Jones, 1980; Owens & MacDonald, 1982). As chronological age increases, to 3-4 years, differences in language performance are reported, not simply an increased slowing down of the acquisition process (Rogers, 1975; Rondal, 1978a; Wiegel-Crump, 1981). It is not clear at this point whether: 1) the deficits associated with advancing age are the result of differences in environmental support for language learning that may decrease with increasing age, 2) there is difficulty in learning the more complex linguistic means to express conceptually complex ideas, or 3) these conceptually complex ideas have not been acquired. Examples of these complex ideas are past, present, and future time concepts; concepts of space and the relation between individuals and objects in space; and concepts of causality (i.e., agent-action-object relations). It is also possible that the outcome of the premature appearance of Alzheimer disease in persons with Down syndrome is being seen (Miniszek, 1983). Only through longitudinal investigation of the language development of children with Down syndrome, including their comprehension and expression of linguistically and cognitively complex notions such as time, will one begin to understand the relation between chronological age, environmental influences, and the cognitive characteristics of Down syndrome.

Vocabulary

The development of vocabulary appears to be independent of other linguistic features in children with Down syndrome. Rondal (1978b) reported that children with Down syndrome have advanced vocabulary diversity over normal children of the same MA and linguistic stage. They use a larger vocabulary when conversing with their mothers in free play than younger normal children in the control sample. Rondal concluded that this lexical advance is present in his subjects because their advanced age reflects more experience, both linguistic and nonlinguistic. He argued that vocabulary, as opposed to other aspects of language, is a direct outcome of environmental experience. Is this advanced development or elaborated development within the same developmental stage? Miller et al. (1978) and Miller et al. (1981) found similar advances in vocabulary in their population of multiply handicapped, and moderately to severely retarded children, who had developmental ages of 7 months-7 years. Apparently, increased vocabulary diversity is not a particular characteristic of children with Down syndrome. It may be a function of experience, as Rondal (1978b) suggested or of language intervention or special education, as postulated by Miller et al. (1981).

Paradoxically, several studies have shown that children with Down syndrome exhibit significant deficits in referential looking behavior, the ability to establish joint reference to objects. Jones (1979, 1980) documented that subjects with Down syndrome, aged 8-22 months developmentally, engaged in less than one-half of the referential looking than their matched normal developmentally subjects. Gunn, Berry, & Andrews (1982) found that the looking behavior of children with Down syndrome had an interpersonal rather than referential quality. Sinson and Wetherick (1982) found differential social interaction patterns which lead to isolation in children with Down syndrome. This was attributable to the subjects' failure to observe the conventions of mutual gaze. These studies together suggest a difference in visual behavior essential to establishing prelinguistic interaction patterns (Schaffer, Collis, & Parsons, 1977); referential communication (Gray, 1978); and language development (Ryan, 1974). Referential looking is a major method of initiating interaction (Jones, 1980) and the establishment of joint reference is essential for vocabulary learning. If these skills are deficient, then why do children with Down syndrome appear to show advanced vocabulary development?

First, vocabulary may be advanced in the number of different words produced relative to total words in a conversation, but may not mark or code more complex conceptual meaning. Second, the data to date are production data only, without corroboration of this population's ability to comprehend advanced meaning or more diverse vocabulary at the same level of conceptual complexity. Third, there is evidence to suggest that lexical and relational meaning may be deficient in subjects with Down syndrome in later periods of development and in subjects who are chronologically older. Several studies have documented that the meanings conveyed by individuals with Down syndrome in conversations are restricted to the here and now (Layton & Sharifi, 1978; Leuder et al., 1981; Rondal, 1978a). Apparently, individuals with Down syndrome fail to transcend time and space in their productive language, talking instead about objects and events immediately present in both time and space.

Comprehension

There are few studies of the language comprehension skills of individuals with Down syndrome, or of the relation between comprehension and production. Several studies have reported on related aspects of comprehension performance.

In two experimental intervention studies, Holdgrafer (1981, 1982) reported variable training rates between comprehension and production for subjects learning plural "s." He interpreted this result as supporting the independence of the two modes in development. Language can develop independently in comprehension and production in children with Down syndrome. This view contrasts sharply with the widely held assumption that comprehension precedes production in language learning (Bloom, 1974; Ingram, 1974). How far comprehension precedes production through the developmental period is not well understood in normal development. In one of the best studies of this relation, Benedict (1979) reported a 5-month gap with comprehension preceding production for the second year of life in a longitudinal study of normal children. Hartley (1982) reported specific comprehension deficits for sequential or syntactic tasks, suggesting right hemisphere dominance for language processing in individuals with Down syndrome. This is in contrast to left hemisphere dominance in normal speakers. The Miller et al. (1981) study documented that comprehension always precedes production, but that production does not always follow the development of comprehension in children with mental retardation. Significant deficits in production relative to comprehension frequently occur.

In two studies of the responses of children with Down syndrome to maternal questions (Leifer & Lewis, 1984; Scherer & Owings, 1984), conflicting results are reported. Scherer & Owings (1984) found that all four of their children with Down syndrome produced appropriate contingent responses to maternal requests for information, clarification, acknowledgment, and action. Correct response rates were 70%-80% for all categories except requests for information, for which two subjects supplied the correct form but not the correct information. They concluded that their subjects perform like normal children of the same linguistic stage. Leifer & Lewis (1984), however, found that the overall response rate of their six children with Down syndrome, four language-matched normal children, and four children matched in age to the language-matched normal group was very low. Appropriate responses to maternal questions were almost nonexistent for the age-matched normal children at 3%, 15% for the language-matched normal children, and 31% for the children with Down syndrome. As well as more frequent appropriate responses, the children with Down syndrome demonstrated advanced responding to directive questions, using more vocalized responses than action responses. Developmentally, action responses precede vocalized responses. Leifer and Lewis (1984) argued that children with Down syndrome show general delays in development compared to chronological age-matched normal children, as well as differences when language level matches are compared. The children with Down syndrome show superior response performance to maternal questions. Leifer and Lewis contended that children with Down syndrome follow a different sequence in the development of syntax and conversational skills than normal children.

Differences in the outcome of these two studies can be attributed to two factors. First, Scherer and Owings (1984) studied children in late stage I (MLU 1.6-2.0) and Leifer and Lewis (1984) studied children at early stage I (MLU 1.0-1.5). Second, the language matches of the Leifer and Lewis study were based on MLU, a production measure, ignoring the fact that children's comprehension abilities show substantial change while they are at the one-word stage of language production (Benedict, 1979; Miller et al., 1980). Leifer and Lewis assumed similar comprehension development based on production ability, an assumption that is not warranted in persons with mental retardation. While production levels were matched, the nature of the response differences reported suggest that the group with Down syndrome had more advanced comprehension skills. Rather than following a different sequence of development, they are simply displaying developmentally advanced responses to maternal questions. This study exemplifies the problem of attempting to match subjects at the one-word level of production on the basis of MLU measures. The one-word period of development extends through most of the second year of life. Matching at this level violates the logic proposed by Brown (1973) for using MLU, that is, as an index of structural development. A recent study by Bless et al. (1985) found that all of their nine subjects with Down syndrome showed delays in both comprehension and production, as compared to their nonverbal cognitive status. These children were functioning between 5 and 8 years developmentally and were tested on measures of vocabulary and syntax. This is the only study of the relations among linguistic variables in both comprehension and production. It confirmed increasing deficits at advanced developmental levels in both comprehension and production. Further, this study documented significant speech-motor problems contributing to the expressive language difficulty of individuals with Down syndrome.

Summary of the Language Characteristics of Children with Down Syndrome

Earlier in this chapter, it was asked if individuals with Down syndrome exhibited specific profiles of language behavior that might be interpreted as evidence for the unique language learning abilities associated with Down syndrome. The answer must be "no" at this time, due primarily to the lack of research on language comprehension for any linguistic characteristic, and the failure of production studies to examine the more advanced aspects of syntax and semantics in favor of more general investigations of pragmatics. Primarily, the review in this chapter has documented the limitations of what can be learned from cross-sectional research. Developmental relations involving changing synchronous interactions require longitudinal study.

Children with Down syndrome show an increasing linguistic deficit in relation to their nonverbal cognitive status with increasing chronological age. This pattern may be the result of specific cognitive deficits associated with the syndrome (including limited capacity for complex representation through language), physiological limitations of perception or motor performance, or a failure of the environment to support language learning through a protracted developmental period. There are few data to support the cognitive deficit view (Cicchetti & Pogge-Hesse, 1982), although three recent investigations have pointed out the increased interest in cognitive, information-processing explanations of the language deficits in persons with Down syndrome. Kahn (1985) found similar cognitive structures in organic (Down syndrome) and non-organic retarded children. Lincoln, Courchesne, Kilman, and Galambos (1985) found, using electrophysiological measures, that children with Down syndrome process some types of information slower than MA- and CA-matched normal children. Ellis, Deacon, and Wooldridge (1985) found evidence for structural memory deficits (in short-term memory) in persons with mental retardation, one-third of whom had Down syndrome. The complexities of tracing specific correlates to language learning becomes more apparent with advancing research.

Physiological links to deficits in language acquisition and use have not been established beyond assumed co-occurrence where deficits in perception or motor performance were found. Studies must be undertaken to uncover profiles of performance, noting change over time.

The last factor, the environment, can be investigated through two sources: studies of maternal language input, and the outcome of language intervention studies. An examination of the literature dealing with the language environment of individuals with Down syndrome follows. It first identifies the factors associated with language facilitation in natural environments and then reports the outcomes of environmental manipulation on language development.

WHAT ARE THE CHARACTERISTICS OF THE LINGUISTIC ENVIRONMENT OF CHILDREN WITH DOWN SYNDROME?

Chapman (1981) reviewed in detail the role of mother-child interaction on language development in the second year of life. She concluded that input plays a demonstrable role in language development when it is specifically contingent upon the child's initiated actions and utterances. She went on to say that it is the linguistically responsive environment, rather than the linguistically stimulating one, that should accelerate language acquisition in the 1- to 2-year-old child. The most important operating principle for the mother seems to be: "Pay attention to what the child is doing and saying" (Chapman, 1981, p. 224).

A number of studies have investigated the assertion that parental expectations due to early knowledge of the condition of Down syndrome affects parent-child interactions. The general pattern that emerges from these studies is as follows. First, both mothers and infants with Down syndrome are less responsive to each other (Stevenson & Leavitt, 1983). Second, mothers of infants with Down syndrome talk to their children more and tend to talk at the same time that their infants are vocalizing (Berger & Cunningham, 1983). Third, mothers of children with Down syndrome speak at a faster rate, producing more utterances per unit time than mothers of normal children (Buck-halt, Rutherford, & Goldberg, 1978). Fourth, mothers of children with Down syndrome are more directive, intrusive, and controlling (Eheart, 1982; Jones, 1979, 1980). Fifth, not all mothers of these children are alike; that is, individual differences in patterns of interaction exist. Thus, maternal sensitivity and directiveness are independent features of maternal behavior (Crawley & Spiker, 1983).

Studies of mothers' speech directed toward their children with Down syndrome have assumed that linguistic complexity is an index of maternal sensitivity to the child's linguistic development. This assumption predicts that maternal language that is judged as too complex or too simple would inhibit language development, while maternal language just above the child's level of linguistic mastery would facilitate language development. The majority of studies have found that maternal language addressed to children with Down syndrome is similar in complexity to that addressed to normal children (Mitchell, 1980; O'Kelley-Collard, 1978; Rondal, 1978a). Two studies, however, produced contradictory results. Mahoney (1983) found that mothers' speech input to young children with Down syndrome did not change over the course of 10 months and was unrelated to the child's behavior. Glaser, Schwethelm, Haffner, and Mahoney (1984) found that mothers of children with Down syndrome produced 79% of their directives toward mother-initiated activities and only 21% of their directives in relation to child-initiated activities. Significantly more directives were successful in changing or facilitating child activities if the activities were child initiated rather than mother initiated. Mothers continued to be successful when they followed the child's lead.

In contrast to the language production literature, the mother-child interaction and maternal language input literature paint a picture of an early lack of maternal responsiveness, followed by a period of adjustment to the child's growing language competence. This view is supported by a study of general responsiveness of mothers to their handicapped children, including children with Down syndrome (Brooks-Gunn & Lewis, 1984). Prior to the onset of intelligible speech, mothers and children with Down syndrome appear mismatched in their communicative interactions. Obviously, these competing pictures (i.e., one of responsiveness and one of lack of responsiveness) should not coexist, if views of normal development are correct. Some insight in the conflicting mother-child literature is provided by Crawley and Spiker (1983), who showed that mothers of children with Down syndrome vary widely in the dimensions of directiveness, sensitivity, and elaborativeness. Mothers who combine sensitivity and directiveness in ways that provide stimulative value may have more competent infants. The maternal characteristic of directiveness has emerged as the most descriptive of mothers of handicapped children. Consider that most intervention programs involve maternal stimulation, some recommending increasing stimulation of all types. It is probable that researchers are recording the results of some of their own advice, correctly or incorrectly interpreted by the mothers of children with Down syndrome.

An increasing awareness of individual variation in the behavior of children with Down syndrome is seen as a function of variation in the child's cognitive and linguistic abilities, hearing level, speech intelligibility and environmental support for language. This chapter now turns to the experimental literature on language intervention with these children as a final window into their language performance.

CAN LANGUAGE INTERVENTION FACILITATE THE LANGUAGE DEVELOPMENT OF CHILDREN WITH DOWN SYNDROME?

Language Intervention

Bricker and Carlson (1981) discussed the convergence of research in cognitive, affective, and linguistic processes in producing new views of language intervention. Intervention programs not accounting for the relation between and among these major domains will be deficient in promoting language as a useful communicative tool. Further, Bricker and Carlson called for an increased emphasis on intervention in the first year of life, prior to the expectation of intelligible speech. Finally, they called on interventionists to exploit the rich interactive feedback networks and reciprocal environmental contexts that exist in the child's environment. In other words, they urged interventionists to do what normal mothers and their children do but to do it more frequently, being particularly responsive to the child's activities and topics, and to the child's communicative needs within his or her environment.

The language intervention literature reviewed in this chapter demonstrates the effectiveness of various teaching techniques. For example, operant conditioning techniques are effective in establishing verbal imitation responses and improving speech intelligibility (Farb & Thorne, 1978; MacCubrey, 1971; Nelson, Peoples, Hay, Johnson, & Hay, 1976). Parent training will not generalize from the laboratory to new contexts (e.g., the home) without instruction (Salzberg & Villani, 1983). Children with Down syndrome show more difficulty in attaining verbal solutions to learning tasks versus nonverbal solutions, which argues the presence of a verbal learning deficit (Rynders, Behlen, & Horrobin, 1979). Training mothers of children with Down syndrome in the use of behavior modification techniques will lead to improved language, self-help skills, and motor and social behavior (Bidder, Bryant, & Gray, 1975). The control of verbal input, lexically and structurally, will improve these children's identification of the target stimuli and improve learning. Also, teacher responsiveness by itself is not sufficient for learning, but must be paired with specification of what is to be learned (Filler, 1976). Together, these studies show that a variety of training techniques can change the language behavior of children with Down syndrome.

Most impressive, however, are the documented effects on children's language development associated with changing maternal interaction style. The approaches to intervention that have had the most productive results are those that have employed developmental interaction models to improve the mother's interaction style, and primarily to improve maternal responsiveness. Seitz (1975) reported on the use of such a model with three mothers of children with Down syndrome. All three children showed gains in language, as documented by an increase in MLU and by an increase in the number of utterances per session, from an average of 42 to 126. The mothers' behavior also significantly changed the children's compliance to the mothers' commands. Compliance increased from 43% to 98% over the 8 weeks of the study. MacDonald, Blott, Gordon, Spiegel, and Hartmann (1974) reported an increase in MLU, and in structural complexity associated with generalization of parent training to new settings. Parents were trained to imitate and expand child utterances. During the last 3 months of the 5-month, study, there were gains equivalent to 3 months of development at the same linguistic stage as Brown's (1973) normal subjects. In a large scale follow-up study, Baker, Heifetz, and Murphy (1980) evaluated 95 children 14 months after parent training. The mothers continued to show growth in the trained skills in three of four areas. Forty-four per cent of the families judged the training to continue to be very useful, and most had incorporated the teaching activities into daily routines.

Ludlow and Alien (1979) conducted a 10-year follow-up of English children who were enrolled in an early intervention program. This program had been introduced to reduce the progressive decline in developmental quotients of children with Down syndrome. All of the 143 children with Down syndrome enrolled in the study were followed. Three groups were investigated. Group A received parent counseling regarding behavior management and techniques to facilitate cognitive, motor, and language development. Group A's program appears to have resembled those of Seitz (1975) and MacDonald et al. (1974), and included periodic visits to a developmental clinic to monitor performance and provide feedback to parents. In group B, the children with Down syndrome were living at home, but the families did not participate in the counseling or developmental clinic programs. In group C, the children with Down syndrome were institutionalized by the second year of life. The group A children showed significantly higher scores than the children in the other two groups on IQ and developmental scales, particularly in personal-social and speech-language development. Mahoney and Snow (1983) investigated cognitive level as a possible predictor of the outcome of language training. Their subjects were 14 children with Down syndrome, ranging in age from 24 to 36 months, and all performed at or below the one-word stage of language production. The experimenters used the MacDonald and Horstmeier (1978) Environmental Language Intervention Program to train parents to facilitate language development. Outcomes of the intervention program were significantly correlated with performance on sensorimotor scales of cognitive development. Children at later sensorimotor stages gained the most and those at earlier stages gained the least from language training.

An Australian study of 36 children with Down syndrome enrolled in an early intervention study that trained mothers to facilitate the language growth of their children reported that all children showed increases in developmental quotients (Clunies-Ross, 1979). Gains were particularly noticeable in social and language areas, with examples cited of children's DQs (developmental quotients) exceeding CA. Younger children made bigger gains than chronologically older children. Finally, Cheseldine and McConkey (1979) analyzed the spontaneous language training strategies of parents. They found that after the mothers were told to teach language, but not how to go about it, three mothers improved their language facilitation strategies and four mothers did not change. Given further opportunity to improve, including direct instruction, two more mothers improved over the control mothers, as evidenced by reduced directiveness and increased responsiveness.

Sign Language Training

The practice of sign language instruction has emerged over the past 10 years in response to children with Down syndrome's failure to learn oral language to the level expected by parents or the educational system. The logic for such instruction is based on two assumptions. First, the child has the cognitive capacity to acquire a representational system. Second, the child has specific deficits that prohibit the acquisition or use of oral language (e.g., a severe speech motor deficit or significant hearing loss). Frequently, however, assumptions are made about the ease of learning a visual language system over spoken language. Such assumptions tend to ignore the fact that both sign language and speech require the same degree of cognitive representational skills, and that both are complete language systems. The issue is complicated by the existence of several sign systems, each of which is fairly independent. Central to the issue, however, is the question of whether teaching sign language facilitates language learning. That is, will children with Down syndrome be able to express themselves more completely using sign language than they could using oral language? Two studies have addressed this question.

Romski and Ruder (1984) investigated the acquisition of agent-object combinations in two conditions: speech alone and speech combined with sign (total communication). The subjects were 10 children with Down syndrome, 3-7 years of age. There were no significant differences found for the two conditions. However, the children exhibited a wide range of individual patterns of development. Romski and Ruder cautioned against automatic adoption or rejection of manual sign systems as a part of oral language training. The second study (Weller, 1981) found that both oral and Signed English training for parents were effective in facilitating language and cognitive growth in 15 children with Down syndrome, 18-36 months of age. Further, they found that the use of signs did not inhibit oral language development. While the oral language versus sign language controversy is far from settled, preliminary data suggest a wide range of individual variation in the skills necessary for language learning. These skills must be carefully evaluated relative to prognosis for language development with either an oral or manual system. The child learning sign is limited to communicating with other signers. The child with no language does not communicate. Signing may enhance oral language development for some children. Signing programs should be considered transitional and augmentative to the development of oral language skills.

Summary

The language intervention studies with children with Down syndrome show that their language behavior can be facilitated with early intervention. Intervention programs emphasizing mothers' language interaction style are particularly effective in increasing language performance. These programs emphasize decreasing the directiveness and complexity of maternal language spoken to the child, increasing responsiveness to the child's initiated behavior (both verbal and nonverbal), and increasing responsiveness to child-initiated topics and the child's communicative needs. Early intervention with families of children with Down syndrome can significantly improve the rate of language development in children with Down syndrome. It appears that a paraphrase of Chapman's (1981) operating principles for mothers of normal children, "Follow the child's lead and respond to their speech and actions," should be adopted by language interventionists as well.

Although early intervention using a directed stimulation model is effective in promoting language learning, there is neither a clear understanding of the specific contingent behaviors leading to language growth nor of how they may change over time. It is the latter of these two points that is particularly troublesome to those concerned with the language development of children with Down syndrome, given the expectation of decreasing language skills over time relative to the development of other skills areas. It is clear that the factors that govern language growth and change in both normal and disordered populations are only beginning to be understood. Clearly there is a great need for further research into the language skills of children with Down syndrome.

QUESTIONS FOR FUTURE RESEARCH

1. Is there developmental continuity in development among linguistic elements, vocabulary, syntax, semantics, and pragmatics, within and between both comprehension and production?
2. Are there individual differences in the patterns of language acquisition that can be explained by characteristics of language input, speech intelligibility, hearing status, or cognitive skills?
3. Can environmental intervention significantly improve language performance? When should intervention begin? Over what period of time is it effective? Can one teach a language system to the child with Down syndrome or can one only provide a supportive environment conducive to language learning?
4. Do early patterns of development predict later performance?

In considering the first question, future research must invoke multidimensional models of speech and language performance. At minimum, these models should include consideration of language comprehension and production; phonology, syntax, semantics, and discourse features; nonverbal cognitive abilities; speech motor performance; and hearing status. The linguistic environment of the child with Down syndrome must also be included. These models must account for developmental change as well as provide for the description of deviant performance at the lexical, utterance, and discourse levels. Such models would provide the bases for hypothesis generation as well as for the evaluation of experimental language interventions.

In order to answer the second, third, and fourth questions, a way to conduct longitudinal research on language change in this population needs to be found. Two major problems must be overcome. First, current funding cycles of 3-5 years preclude studying anything more than 1 year of development. Second, many children with Down syndrome in the United States are in some type of early intervention/special education program. Mechanisms must be developed to bring the researcher into the intervention/special education circle so they can learn which practices work and which ones do not. An example of a productive strategy can be found in Gibson and Ingram (1983). In this study, a parent and a researcher combined efforts to conduct a longitudinal study of a child with delayed language. The result is an impressive study with a very rich data base. Parents, "bring a researcher home to dinner." Researchers, listen to what parents have to say, and take heed of their concerns. The goal is the same: improved communication for children with Down syndrome.

Longitudinal research demands long-term commitment on the part of researchers, requiring at least 10 years to study 5 years of longitudinal development. Most researchers consider this very high risk research, given that one-third to one-fourth of a career may be at stake. Funding mechanisms and support networks must be developed to form communities of scholars who will work jointly with families of children with Down syndrome.

New tools will have to be designed and implemented to overcome the data analysis problems inherent in language research. The development of computer analysis techniques may provide the significant advances necessary to make longitudinal research possible. Computers can be used to analyze free speech samples, as well as to gather data on language comprehension (Miller & Chapman, 1984). Computer formats for free speech sample analysis provide a standard for transcription that can be read by clinicians and researchers alike. Data exchanges for language data of various populations must be encouraged (MacWhinney & Snow, 1985). Longitudinal research will require a great deal of data and data analysis. The use of computers for exchange and analysis of the data will finally bring true longitudinal study of communication development in children with Down syndrome into the realm of reality.

REFERENCES

Baker, B. L., Heifetz, L. J., & Murphy, D. M. (1980). Behavioral training for parents of mentally retarded children: One-year follow-up. American Journal of Mental Deficiency, 85, 31-38.

Bates, E., & MacWhinney, B. (1979). A functionalist approach to the acquisition of grammar. In E. Ochs & B. Schieffelin (Eds.), Developmental pragmatics (pp. 167-214). New York: Academic Press.

Baumeister, A. (1967). Problems in comparative studies of mental retardates and normals. American Journal of Mental Deficiency, 71, 869-875.

Baumeister, A. (1984). Some methodological and conceptual issues in study of cognitive processes with retarded people. In P. Brooks, R. Sperber, & C. McCauley (Eds.), Learning and cognition in the mentally retarded (pp. 6-38). Hillsdale, NJ: Lawrence Erlbaum Associates.

Benedict, H. (1979). Early lexical development: Comprehension and production. Journal of Child Language, 6, 183-200.

Berger, J., & Cunningham, C. C. (1983). Development of early vocal behaviors and interactions in Down's syndrome and nonhandicapped infant-mother pairs. Developmental Psychology, 19, 322-331.

Bidder, R. T., Bryant, G., & Gray, O. P. (1975). Benefits to Down's syndrome children through training their mothers. Archives of Disease in Childhood, 50, 383-386.

Bless, D., Swift, E., & Rosen, M. (1985). Communication profiles of children with Down syndrome. Unpublished manuscript, University of Wisconsin, Waisman Center on Mental Retardation and Human Development, Madison.

Bloom, L. (1974). Talking, understanding, and thinking. In L. Lloyd (Ed.), Language perspectives: Acquisition, retardation, and intervention (pp. 285-312). Baltimore: University Park Press.

Brandes, P., & Elsinger, D. (1981). The effects of early middle ear pathology on auditory perception and academic achievement. Journal of Speech and Hearing Disorders, 46, 301-307.

Bricker, D., & Carlson, L. (1981). Issues in early language intervention. In R. Schiefelbusch & D. Bricker (Eds.), Early language: Acquisition and intervention (pp. 477-516). Baltimore: University Park Press.

Brooks-Gunn, J., & Lewis, M. (1984). Maternal responsivity in interactions with handicapped infants. Child Development, 55, 782-793.

Brown, R. (1973). A first language. Cambridge, MA: Harvard University Press.

Buckhalt, J. A., Rutherford, R. B., & Goldberg, K. E. (1978). Verbal and nonverbal interaction of mothers with their Down's syndrome and nonretarded infants. American Journal of Mental Deficiency, 82, 337-343.

Chapman, R. (1981). Mother-child interaction in the second year of life: Its role in language development. In R. Schiefelbusch & D. Bricker (Eds.), Early language: Acquisition and intervention (pp. 201-250). Baltimore: University Park Press.

Chapman, R., Miller, J., MacKenzie, H., & Bedrosian, J. (1981, August). The development of discourse skills in the second year of life. Paper presented at the 2nd International Congress for the Study of Child Language, University of British Columbia, Vancouver, British Columbia, Canada.

Cheseldine, S., & McConkey, R. (1979). Parental speech to young Down's syndrome children: An interview study. American Journal of Mental Deficiency, 83, 612-620.

Cicchetti, D., & Pogge-Hesse, P. (1982). Possible contributions of the study of organically retarded persons to developmental theory. In E. Zigler & D. Balla (Eds.), Mental retardation: The developmental-difference controversy (pp. 277-318). Hillsdale, NJ: Lawrence Erlbaum Associates.

Clunies-Ross, G. G. (1979). Accelerating the development of Down's syndrome infants and young children. The Journal of Special Education, 13, 169-177.

Coggins, T. E. (1979). Relational meaning encoded in the two-word utterances of stage 1 Down's syndrome children. Journal of Speech and Hearing Research, 22, 166-178.

Coggins, T. E., & Carpenter, R. L. (1981). The communicative intention inventory: A system for observing and coding children's early intentional communication. Applied Psycholinguistics, 2, 235-251.

Coggins, T. E., Carpenter, R. L., & Owings, N. O. (1983). Examining early intentional communication in Down's syndrome and nonretarded children. British Journal of Disorders of Communication, 18, 98-106.

Coggins, T. E., & Stoel-Gammon, C. (1982). Clarification strategies used by four Down's syndrome children for maintaining normal conversational interaction. Education and Training of the Mentally Retarded, 17, 65-67.

Crawley, S. B., & Spiker, D. (1983). Mother-child interactions involving two-year-olds with Down syndrome: A look at individual differences. Child Development, 54, 1312-1323.

Cromer, R. (1981). Reconceptualizing language acquisition and cognitive development. In R. Schiefelbusch & D. Bricker (Eds.), Early language: Acquisition and intervention (pp. 51-138). Baltimore: University Park Press.

deVilliers, P., & deVilliers, J. (1985). The acquisition of English. In D. Slobin (Ed.), The cross-linguistic study of language acquisition. Hillsdale, NJ: Lawrence Erlbaum Associates.

Downs, M. (1980). The identification of children at risk for middle ear effusion problems. Annals of Otology, Rhinology and Laryngology, 89(3, Part 2), 168-171.

Duchan, J., & Erickson, J. (1976). Normal and retarded children's understanding of semantic relations in different verbal contexts. Journal of Speech and Hearing Research, 19, 767-776.

Eheart, B. K. (1982). Mother-child interactions with nonretarded and mentally retarded preschoolers. American Journal of Mental Deficiency, 87, 20-25.

Ellis, N. R., Deacon, J. R., & Wooldridge, P. W. (1985). Structural memory deficits of mentally retarded persons. American Journal of Mental Deficiency, 89, 393-402.

Evans, D. (1977). The development of language abilities in mongols: A correlational study. Journal of Mental Deficiency Research, 21, 103-117.

Farb, J., & Thorne, J. (1978). Improving the generalized mnemonic performance of a Down's syndrome child. Journal of Applied Behavior Analysis, 11, 413-419.

Filler, J. W., Jr. (1976). Modifying maternal teaching style: Effects of task arrangement on the match-to-sample performance of retarded preschool-age children. American Journal of Mental Deficiency, 80, 602-612.

Finch-Williams, A. (1984). The developmental relationship between cognition and communication: Implications for assessment. Topics in Language Disorders, 5, 1-13.

Gibson, D., & Ingram, D. (1983). The onset of comprehension and production in a language delayed child. Applied Psycholinguistics, 4, 359-376.

Glaser, K., Schwethelm, B., Haffner, P., & Mahoney, G. (1984, May). The effects of maternal directives on the performance of Down syndrome infants. Paper presented at the Annual Conference of the American Association on Mental Deficiency, Minneapolis.

Graham, J., & Graham, L. (1971). Language behavior of the mentally retarded: Syntactic characteristics. American Journal of Mental Deficiency, 75, 623-629.

Gray, H. (1978). Learning to take an object from the mother. In A. Lock (Ed.), Action, gesture, and symbol: The emergence of language. New York: Academic Press.

Greenwald, C. A., & Leonard, L. B. (1979). Communicative and sensorimotor development of Down's syndrome children. American Journal of Mental Deficiency, 84, 296-303.

Gunn, P., Berry, P., & Andrews, R. J. (1982). Looking behavior of Down syndrome infants. American Journal of Mental Deficiency, 87, 344-347.

Harris, J. (1983). What does mean length of utterance mean? Evidence from a comparative study of normal and Down's syndrome children. British Journal of Disorders of Communication, 18, 153-169.

Hartley, X. (1982). Receptive language processing by Down's syndrome children. Journal of Mental Deficiency Research, 26, 263-269.

Holdgrafer, G. (1981). Mode-relations in language learning by language-deficient retarded subjects. Perceptual and Motor Skills, 53, 520-522.

Holdgrafer, G. (1982). Teaching comprehension and production. Perceptual and Motor Skills, 55, 306.

Ingram, D. (1974). Comprehension and production. In R. Schiefelbusch & L. Lloyd (Eds.), Language perspectives: Acquisition, retardation, and intervention. Baltimore: University Park Press.

Jones, O. (1979). A comparative study of mother-child communication with Down syndrome and normal infants. In D. Schaffer & J. Dunn (Eds.), The first year of life (pp. 175-195). New York: John Wiley & Sons.

Jones, O. H. M. (1980). Prelinguistic communication skills in Down's syndrome and normal infants. In T. F. Field (Ed.), High-risk infants and children: Adult and peer interactions (pp. 205-225). New York: Academic Press.

Kahn, J. (1985). Evidence of the similar-structure hypothesis controlling for organicity. American Journal of Mental Deficiency, 89, 372-378.

Kamhi, A., & Johnston, J. (1982). Towards an understanding of retarded children's linguistic deficiencies. Journal of Speech and Hearing Research, 25, 435-445.

Lackner, J. R. (1968). A developmental study of language behavior in retarded children. Neuropsychologia, 6, 301-320.

Layton, T. L., & Sharifi, H. (1978). Meaning and structure of Down's syndrome and nonretarded children's spontaneous speech. American Journal of Mental Deficiency, 83, 439-445.

Lee, L. (1974). Developmental sentence analysis. Evanston, IL: Northwestern University Press,

Leifer, J., & Lewis, M. (1984). Acquisition of conversational response skills by young Down Syndrome and nonretarded children. American Journal of Mental Deficiency, 88, 610-618.

Lenneberg, E., Nichols, I., & Rosenberger, E. (1964). Primitive stages of language development in mongolism. Research Publications: Association for Research in Nervous and Mental Disease, 42.

Leonard, L. (1978). Cognitive factors in early linguistic development. In R. Schiefelbusch (Ed.), Bases of language intervention. Baltimore: University Park Press.

Leuder, L, Fraser, W. L, & Jeeves, M. A. (1981). Social familiarity and communication in Down syndrome. Journal of Mental Deficiency Research, 25, 133-142.

Lincoln, A. J., Courchesne, E., Kilman, B. A., & Galambos, R. (1985). Neurophysiological correlates of information-processing by children with Down syndrome. American Journal of Mental Deficiency, 89, 403-414.

Ludlow, J. R., & Alien, L. M. (1979). The effect of early intervention and preschool stimulus on the development of the Down's syndrome child. Journal of Mental Deficiency Research, 23, 29-44.

MacCubrey, J. (1971). Verbal operant conditioning with young institutionalized Down's syndrome children. American Journal of Mental Deficiency, 75, 696-701.

MacDonald, J. D., Blott, J. P., Gordon, K., Spiegel, B., & Hartmann, M. (1974). An experimental parent-assisted treatment program for preschool language-delayed children. Journal of Speech and Hearing Disorders, 39, 395-415.

MacDonald, J., & Horstmeier, D. (1978). Environmental Language Intervention Program. Columbus, OH: Charles E. Merrill.

MacWhinney, B. (1984). Child Language Data Exchange System (CHILDES), Mac-Arthur Foundation Project. Pittsburgh: Carnegie-Mellon University.

MacWhinney, B., & Snow, C. (1985). The child language data exchange. Journal of Child Language, 12, 271-296.

Mahoney, G. (1983). A developmental analysis of communication between mothers and infants with Down's syndrome. Topics in Early Childhood Special Education, 3, 63-76.

Mahoney, G., & Snow, K. (1983). The relationship of sensorimotor functioning to children's response to early language training. Mental Retardation, 21, 248-254.

McCarthy, D. (1954). Language development in children. In L. Carmichael (Ed.), Manual of child psychology (pp. 492-630). New York: John Wiley & Sons.

Miller, J. (1981). Analyzing language production in children: Experimental procedures. Baltimore: University Park Press.

Miller, J., & Chapman, R. (1984). Disorders of communication: Investigating the development of language of mentally retarded children. American Journal of Mental Deficiency, 88, 536-545.

Miller, J., Chapman, R., & Bedrosian, J. (1978). The relationship between etiology, cognitive development, and language and communicative performance. The New Zealand Speech Therapists' Journal, 33, 2-11.

Miller, J., Chapman, R. S., Branston, M. B., & Reichle, J. (1980). Language comprehension in sensorimotor stages V and VI. Journal of Speech and Hearing Research, 23, 284-311.

Miller, J., Chapman, R., & MacKenzie, H. (1981). Individual differences in the language acquisition of mentally retarded children. Paper presented at the 2nd International Congress for the Study of Child Language, Vancouver, British Columbia, Canada.

Miller, J., & Yoder, D. (1974). An ontogenetic language teaching strategy for retarded children. In R. Schiefelbusch & L. Lloyd (Eds.), Language perspectives: Acquisition, retardation, and intervention (pp. 505-528). Baltimore: University Park Press.

Miniszek, N. A. (1983). Development of Alzheimer disease in Down syndrome individuals. American Journal of Mental Deficiency, 87, 377-385.

Mitchell, D. R. (1980). Down's syndrome children in structured dyadic communication situations with their parents. In J. Hogg & P. J. Mittler (Eds.), Advances in mental handicap research (Vol. 2, pp. 161-194). New York: John Wiley & Sons.

Naremore, R., & Dever, R. (1975). Language performance of educable mentally retarded and normal children at five age levels. Journal of Speech and Hearing Research, 18, 82-96.

Nelson, R. O., Peoples, A., Hay, L. R., Johnson, T., & Hay, W. (1976). The effectiveness of speech training techniques based on operant conditioning: A comparison of two methods. Mental Retardation, June, 34-38.

Newfield, M., & Schlanger, B. (1968). The acquisition of English morphology by normals and educable mentally retarded children. Journal of Speech and Hearing Research, 11, 693-706.

Nisbet, J., Zanella, K., & Miller, J. (1984). An analysis of conversations among handicapped students and a nonhandicapped peer. Exceptional Children, 51, 156-162.

O'Kelley-Collard, M. (1978). Maternal linguistic environment of Down's syndrome children. The Australian Journal of Mental Retardation, 5, 121-126.

Owens, R. E., Jr., & MacDonald, J. D. (1982). Communicative uses of the early speech of nondelayed and Down syndrome children. American Journal of Mental Deficiency, 86, 503-510.

Piaget, J., & Inhelder, B. (1969). The psychology of the child. New York: Basic Books.

Price-Williams, D., & Sabsay, S. (1979). Communicative competence among severely retarded persons. Simiotica, 26, 35-63.

Rogers, M. G. H. (1975). A study of language skills in severely subnormal children. Child: Care, Health and Development, 1, 113-126.

Romski, M. A., & Ruder, K. F. (1984). Effects of speech and speech and sign instruction on oral language learning and generalization of action + object combinations by Down's syndrome children. Journal of Speech and Hearing Disorders, 49, 293-302.

Rondal, J. A. (1978a). Developmental sentence scoring procedure and the delay-difference question in language development of Down's syndrome children. Mental Retardation, April, 169-171.

Rondal, J. (1978b). Maternal speech to normal and Down's syndrome children matched for mean length of utterance. In C. E. Meyers (Ed.), Quality of life in severely and profoundly mentally retarded people: Research foundations for improvement (pp. 193-266). Washington, DC: American Association on Mental Deficiency.

Ryan, J. (1974). Early language development: Towards a communicational analysis. In M. Richards (Ed.), The integration of the child into a social world. London: Cambridge University Press.

Ryan, J. (1975). Mental subnormality and language development. In E. Lenneberg & E. Lenneberg (Eds.), Foundations of language development: A multidisciplinary approach (Vol. 2, pp. 269-278). New York: Academic Press.

Rynders, J. E., Behlen, K. L., & Horrobin, J. M. (1979). Performance characteristics of preschool Down's syndrome children receiving augmented or repetitive verbal instruction. American Journal of Mental Deficiency, 84, 67-73.

Salzberg, C. L., & Villani, T. V. (1983). Speech training by parents of Down syndrome toddlers: Generalization across settings and instructional contexts. American Journal of Mental Deficiency, 87, 403-413.

Sanger, R. G. (1975). Facial and oral manifestations of Down's syndrome. In M. Koch & F. de la Cruz (Eds.), Down's syndrome (Mongolism): Research, prevention and management (pp. 32-46). New York: Brunner/Mazel.

Schaffer, H., Collis, G., & Parsons, G. (1977). Vocal interchange and visual regard in verbal and pre-verbal children. In H. R. Schaffer (Ed.), Studies in mother-infant interaction. New York: Academic Press.

Scherer, N. J., & Owings, N. O. (1984). Learning to be contingent: Retarded children's responses to their mothers' requests. Language and Speech, 27, 255-267.

Seitz, S. (1975). Language intervention-Changing the language environment of the retarded child. In M. Koch & F. de la Cruz (Eds.), Down's syndrome (Mongolism): Research, prevention and management (pp. 157-179). New York: Brunner/Mazel.

Semmel, M., Barritt, L., Bennett, S., & Perfetti, C. (1967). The performance of educable mentally retarded and normal children on a modified cloze task. Studies in Language Learning and Language Behavior, 5, 326-342.

Sinson, J. C., & Wetherick, N. E. (1982). Mutual gaze in pre-school Down's and normal children. Journal of Mental Deficiency Research, 26, 123-129.

Slobin, D. (1985). Cross-linguistic evidence for language-making capacity. In D. Slobin (Ed.), The cross-linguistic study of language acquisition. Hillsdale, NJ: Lawrence Erlbaum Associates.

Smith, B., & Oiler, K. (1981). A comparative study of pre-meaningful vocalizations produced by normally developing and Down's syndrome infants. Journal of Speech and Hearing Research, 46, 46-51.

Smith, M. (1941). Measurement of the size of general English vocabulary through the elementary grades and high school. Genetic Psychology Monographs, 24, 311-345.

Snyder, L. (1978). Communication and cognitive abilities and disabilities in the sensorimotor period. Merrill-Palmer Quarterly, 24, 161-180.

Snyder, L. K., & McLean, J. E. (1976). Deficient acquisition strategies: A proposed conceptual framework for analyzing severe language deficiency. American Journal of Mental Deficiency, 81, 338-349.

Stevenson, M. B., & Leavitt, L. A. (1983). Mother-infant interaction: A Down's syndrome case study. Paper presented at the 2nd International Workshop on the "At Risk" Infant, Jerusalem.

Stoel-Gammon, L. (1980). Phonological analysis of four Down's syndrome children. Applied Psycholinguistics, 1, 31-48.

Weller, E. (1981). Comparison of oral and signed-English communication training with Down's syndrome children in a parent-assisted language intervention program. Dissertation Abstracts International, (University Microfilms).

Wells, G. (1980). Apprenticeship in meaning. In K. Nelson (Ed.), Children's Language (Vol. 2). New York: Gardner Press.

Wiegel-Crump, C. A. (1981). The development of grammar in Down's syndrome children between the mental ages of 2-0 and 6-11 years. Education and Training of the Mentally Retarded, February, 24-30.

Yoder, D., & Miller, J. (1972). What we may know and what we can do: Input toward a system. In J. McLean, D. Yoder, & R. Schiefelbusch (Eds.), Language intervention with the retarded: Developing strategies (pp. 89-110). Baltimore: University Park Press.

The writing of this chapter was supported in part by Core Research Support to the Waisman Center on Mental Retardation and Human Development from NICHD, Grant #2 P30 HDO3352, 1985. The author would like to express his thanks to Gloria Streit, Pam Laikko, John Rynders, and Robin Chapman for their critical comments.