Relationship between theory of mind and language ability in children and adolescents with intellectual disability.

BACKGROUND: The present study was designed to evaluate the validity of the false belief task as a measure of theory of mind development in individuals with intellectual disability (ID). In most if it variants, the false belief task is linguistically demanding. This raises the possibility that the finding that individuals with ID do poorly on it might reflect language difficulties rather than theory of mind difficulties. Complicating matters further, however, is the fact that there are theoretical reasons to suppose that there might be a relationship between some dimensions of language ability and theory of mind development in individuals with ID (as well as in other populations). METHOD: In the present study, children and adolescents with ID and typically developing (non-verbal) mental age matches completed a standard false belief task and several tasks designed to measure language ability. RESULTS: We reasoned that a pattern in which false belief performance was correlated with all measures of language ability would reflect an artefactual relationship, whereas a more highly circumscribed, theoretically sensible pattern of correlations that was similar across both groups would support the validity of the false belief task. CONCLUSIONS: The results indicated that for individuals with ID who have limited narrative language skills, those limitations contribute substantially to their failure on the false belief task. For individuals with ID who have more highly developed narrative language skills (about 40% of the sample tested), however, the false belief task may provide a valid measure of their progress towards acquiring an adequate theory of mind. This latter conclusion was suggested by the fact screening out individuals who failed to meet linguistic and cognitive prerequisites for dealing with the performance demands of the false belief task yielded non-significant correlations between false belief performance and the language measures for both the group with ID and the typically developing comparison group.

Parental scaffolding of the discourse of children and adolescents with intellectual disability: the case of, referential expressions.

Individuals with intellectual disability find the process of establishing referents to be an especially challenging component of discourse. The present study was designed to examine whether these problems partly result from a failure of parents to appropriately scaffold the discourse participation of individuals with intellectual disability. Children and adolescents with intellectual disability and their parents participated in two dyadic non-face-to-face referential tasks which afforded parents an opportunity to scaffold their children's behaviour as speaker and as listener. Comparisons were made with parents and their typically developing children who completed the same tasks. It was found that the parents of individuals with intellectual disability scaffolded their children's discourse participation to the same extent, as effectively and in the same manner as the parents of the typically developing children. The former were also found to adjust their scaffolding according to their children's level of behavioural competence. In summary, there is no evidence that parents contribute to the referential problems of individuals with intellectual disability.

Understanding referential expressions in context: use of common ground by children and adolescents with mental retardation.

Listeners interpret utterances against the common ground, or network of presuppositions shared with the speaker. The first purpose of the study was to determine whether individuals with mental retardation use the major sources of common ground (i.e., physical copresence, linguistic copresence, and community membership) to resolve referential ambiguity. The second purpose was to determine whether they seek confirmation of their referent choices in accordance with the certainty of interpretation afforded by the common ground. The third purpose was to determine whether they signal noncomprehension when faced with ambiguity and common ground that is not informative. The final purpose was to evaluate the relationship between within-group variability in common ground use and measures of nonverbal cognition, receptive and expressive language, and social cognition. Participants were school-age individuals with mental retardation and typically developing children matched to them on nonverbal MA. Common ground use was examined in a role-playing task in which the participant responded to ambiguous utterances. Common ground was manipulated within participants. We determined whether referent selections were appropriate for the common ground, whether they were accompanied by confirmation requests, and whether noncomprehension was signaled. Both groups used all sources of common ground to resolve referential ambiguity at better than chance levels but were less successful in using community membership. Both groups also requested confirmation of their referent choices most often when the common ground was based on community membership. Both groups signaled noncomprehension when the common ground was not informative. Different aspects of common ground use were related to different predictors for the group with mental retardation.

Preschoolers' communication during scripted interactions.

The communicative interactions of 15 dyads of four- to five-year-olds during pretend play involving routine, or scripted, events were investigated as a function of the children's knowledge of the scripts. Measures of the quantity and quality of interaction and the strategies that the children used to establish mutual knowledge (i.e. assess and adapt to their discourse partner's level of expertise), which is essential to good communication, were examined. Each dyad participated in a MATCHED condition (both members had extensive knowledge of the script) and a MISMATCHED condition (one member had extensive script knowledge and the other did not). Shared script knowledge facilitated communicative interactions, as indicated by more topic maintenance and fewer requests for clarification in the matched condition than in the mismatched condition. The children attempted to establish mutual knowledge more frequently in the mismatched condition than in the matched condition and, moreover, mutual knowledge establishment was related to the children's communicative effectiveness.

Signaling of noncomprehension by children and adolescents with mental retardation: effects of problem type and speaker identity.

Previous research has demonstrated considerable within-individual and within-group variability in the signaling of noncomprehension by persons with mental retardation. The first purpose of this study was to determine whether within-individual variability in such signaling was related to differences in the nature of the inadequate message and the identity of the speaker. The second purpose was to evaluate the relationship between within-group variability in noncomprehension signaling and measures of cognition, receptive and excessive language ability, speech intelligibility, and social cognition. Participants were school-age individuals with mild mental retardation and typically developing children matched to them on nonverbal MA. Noncomprehension signaling was examined in a direction-following task in which inadequate message type and speaker were manipulated. It was found that message type, but not speaker, influenced noncomprehension signaling, with no difference between the two groups. We also found that performance on a test of receptive language ability was the best predictor of noncomprehension signaling for persons with mental retardation.