Use of speaker intent and grammatical cues in fast-mapping by adolescents with Down syndrome.

PURPOSE: The authors asked whether adolescents with Down syndrome (DS) could fast-map novel nouns and verbs when word learning depended on using the speaker's pragmatic or syntactic cues. Compared with typically developing (TD) comparison children, the authors predicted that syntactic cues would prove harder for the group with DS to use and that action verbs would be harder to fast-map than nouns. METHOD: Twenty participants with DS, aged 12-18 years, and 19 TD participants, aged 3-6 years, were matched on syntax comprehension and engaged in 4 fast-mapping tasks. Both comprehension and production of novel words were assessed for each task. Through use of hierarchical regression models, hearing, cognition, and working memory were considered as predictors of total comprehension and production performance for each group. RESULTS: Both groups used speaker intent in fast-mapping labels to object referents but performed more poorly on fast-mapping verbs. Neither group appeared to use grammatical cues to disambiguate the intended referent. Syntax comprehension was replicated as a predictor of fast-mapping comprehension for the DS group. Syntax comprehension and chronological age were replicated as predictors of TD fast-mapping. CONCLUSION: Participants with DS had better recall for the object or action seen most recently during speaker intent tasks. They had better recall for the object presented first and the action presented last during grammatical cue tasks. Response patterns may have depended on the structure of specific task paradigms. Verb acquisition may be facilitated when either the action or label is familiar.

Language learning in Down syndrome: the speech and language profile compared to adolescents with cognitive impairment of unknown origin.

Children and adolescents with Down syndrome show an emerging profile of speech and language characteristics that is typical of the syndrome (Chapman & Hesketh, 2000; Chapman, 2003; Abbeduto & Chapman, 2005) and different from typically developing children matched for nonverbal mental age, including expressive language deficits relative to comprehension that are most severe for syntax, and, in adolescence, strengths in comprehension vocabulary, improvements in expressive syntax, but losses in comprehension of syntax (Chapman, Hesketh & Kistler, 2002). Here we compare 20 adolescents with Down syndrome to 16 individuals with cognitive impairment of unknown origin, statistically matched for age and nonverbal mental age, to show that the age-related strengths in vocabulary comprehension are not limited to the Down syndrome phenotype, but are limited to a certain type of vocabulary test: for both groups, performance on the Peabody Picture Vocabulary Test-3 is significantly greater than performance on the vocabulary subtest of the Test of Auditory Comprehension of Language-3, which does not differ from the syntax comprehension subtests. Vocabulary size, but not conceptual level, is a strength for adolescents with cognitive impairment. In contrast, deficits in auditory-verbal working memory, syntax and vocabulary comprehension, and narration of picture-books without an opportunity to preview them are all specific to the adolescent group with Down syndrome. The expressive language deficit disappears when a preview opportunity and picture support is given.

Effect of memory support and elicited production on fast mapping of new words by adolescents with Down syndrome.

PURPOSE: The purpose of this study was to determine whether memory support and elicited production differentially benefited fast mapping of new vocabulary (comprehension, production accuracy, and speed) in adolescents with Down syndrome (DS) compared with typically developing (TD) children matched for syntax comprehension. The study also examined predictors of individual difference within groups. METHOD: The ability of 19 adolescents with DS to fast map new noun vocabulary was compared with 18 TD children matched for syntax comprehension. The effects of memory support (5 examiner repetitions of the novel word vs. 1) and elicited production (2 repetitions by child vs. none) were evaluated in a repeated-measures design with counterbalancing of tasks. RESULTS: For novel word comprehension, low memory support impaired the TD but not DS participants. DS participants were slower to respond correctly in all conditions but benefited more from memory support than TD participants in speed of response. For novel word production, high memory support benefited all participants in the no-elicited-production condition. Elicited production improved performance equally in the 2 memory conditions. Standard comprehension measures predicted fast-mapped comprehension: vocabulary in the case of the DS group, syntax in the case of the TD group. Auditory short-term memory measures predicted fast-mapped production in the TD group but not the DS group. Hearing and grammatical morpheme comprehension predicted novel word production in the DS group. CONCLUSIONS: DS participants' speed of comprehension of fast-mapped words was differentially increased compared with the TD syntax-comprehension-matched group, by memory support. Elicited production improved production fast-mapping for all participants, and memory support improved it in the absence of elicited production. Comprehension predicted individual differences in fast-mapped comprehension in DS; in addition, hearing predicted production.

Sentence comprehension in adolescents with Down syndrome and typically developing children: role of sentence voice, visual context, and auditory-verbal short-term memory.

The authors evaluated the roles of auditory-verbal short-term memory, visual short-term memory, and group membership in predicting language comprehension, as measured by an experimental sentence comprehension task (SCT) and the Test for Auditory Comprehension of Language--Third Edition (TACL-3; E. Carrow-Woolfolk, 1999) in 38 participants: 19 with Down syndrome (DS), age 12 to 21 years, and 19 typically developing (TD) children, age 3 to 5 years, matched on syntax comprehension, as measured by TACL-3 Subtests II and III. Of the 5 dependent measures of comprehension, auditory-verbal short-term memory accounted for significant amounts of variance in 4; group membership, 1 (semantic role assignment); and visual short-term memory, 0. In the group with DS, hearing status predicted variation in Grammatical Morphemes (TACL-3 Subtest II). Using the SCT, the authors also investigated the effects of varying sentence voice and supporting visual context on sentence comprehension. SCT performance was significantly poorer in terms of (a) referent selection and semantic role assignment, for passive (vs. active) sentences in both groups, and (b) semantic role assignment in all sentences for the group with DS (vs. the TD group). Vocabulary strengths in the group with DS were found with the Peabody Picture Vocabulary Test--Third Edition (L. M. Dunn & L. M. Dunn, 1997) but not the TACL-3 Vocabulary subtest.

Fast mapping of words and story recall by individuals with Down syndrome.

This study compared adolescents with Down syndrome to nonverbal mental-age matched controls in their ability to fast map new noun vocabulary in spoken story contexts. Context for novel words varied within subjects in the distance between mentions (close-distant) and the ease of inferring a real word for the referent (specificity). The 23 participants with Down syndrome (DS) were aged 12.8-20.3 years. The 24 typically developing (TD) children, matched on visual nonverbal mental age (MA), were 4.1 to 6.1 years old. Participants listened to 4 tape-recorded stories, each containing 3 mentions of 2 novel words in close or distant proximity and with clear or uncertain reference, and recalled each story after presentation. Fast-mapping production was measured by the occurrence of the novel word in story recall. Fast-mapping comprehension was measured by asking children to define the novel words. The DS group did not differ from the TD group in novel word production but seemed to have more difficulty with novel word definition. For both groups, novel word production was higher in the nonspecific than the specific referent condition, suggesting that availability of a real word label interfered with fast mapping. Recall of story propositions was poorer for the DS group. For both groups, story recall was better for text units not directly associated with novel words than for text units containing novel words, suggesting a trade-off effect in processing. Regression analyses indicated that syntax comprehension, rather than mean length of utterance, predicted novel word production in both groups; MA additionally contributed to predict DS story recall.

Predicting longitudinal change in language production and comprehension in individuals with Down syndrome: hierarchical linear modeling.

Longitudinal change in syntax comprehension and production skill, measured four times across a 6-year period, was modeled in 31 individuals with Down syndrome who were between the ages of 5 and 20 years at the start of the study. Hierarchical Linear Modeling was used to fit individual linear growth curves to the measures of syntax comprehension (TACL-R) and mean length of spontaneous utterances obtained in 12-min narrative tasks (MLU-S), yielding two parameters for each participant's comprehension and production: performance at study start and growth trajectory. Predictor variables were obtained by fitting linear growth curves to each individual's concurrent measures of nonverbal visual cognition (Pattern Analysis subtest of the Stanford-Binet), visual short-term memory (Bead Memory subtest), and auditory short-term memory (digit span), yielding two individual predictor parameters for each measure: performance at study start and growth trajectory. Chronological age at study start (grand-mean centered), sex, and hearing status were also taken as predictors. The best-fitting HLM model of the comprehension parameters uses age at study start, visual short-term memory, and auditory short-term memory as predictors of initial status and age at study start as a predictor of growth trajectory. The model accounted for 90% of the variance in intercept parameters, 79% of the variance in slope parameters, and 24% of the variance at level 1. The some predictors were significant predictors of initial status in the best model for production, with no measures predicting slope. The model accounted for 81% of the intercept variance and 43% of the level 1 variance. When comprehension parameters are added to the predictor set, the best model, accounting for 94% of the intercept and 22% of the slope variance, uses only comprehension at study start as a predictor of initial status and comprehension slope as a predictor of production slope. These results reflect the fact that expressive language acquisition continues in adolescence and is predicted by syntax comprehension and its growth trajectory.

Narrative content as described by individuals with Down syndrome and typically developing children.

Narratives of the wordless picture story, Frog, Where Are You?, by 33 individuals with Down syndrome and typically developing children (33 matched for mental age, 33 for syntax comprehension, 33 for mean length of utterance) were analyzed for expression of plot line, story theme, and the protagonists' misadventures in the story. Despite their restricted expressive syntax and vocabulary, the group with Down syndrome expressed more plot line and thematic content and more of one of the protagonists' misadventures than the MLU controls; they most resembled the syntax comprehension control participants. We conclude that the group with Down syndrome had a conceptual understanding of the picture story similar to that of the TACL-R group and a strategy for expressing that understanding despite expressive lexical and syntactic limitations; this resulted in the expression of more narrative content than formal measures of expressive language would predict. We propose that the higher syntactic comprehension skills of the group with Down syndrome, combined with their experience with story content (listening to stories), may have contributed to their developing higher-level story schemas than would be expected given their MLUs.