Implicit learning of relative clause in Cantonese-speaking children with and without developmental language disorder

Abstract

Much of the research on Development Language Disorder (DLD) have reported on the causes of the language problems observed in the affected children. Explanations from neurobiological, linguistic and cognitive perspectives have offered solid though inconclusive understanding of the complexity of the causes that underly DLD. Much research on the causes focused on the language problems observed but not the language learning problem itself. Language is complex and explanations for DLD are likely to be multi-factorial. The search of a multi-factorial explanation of DLD will benefit from a more direct and complementary investigation of the children’s language learning in action. Structural priming allows systematic investigations of children’s implicit learning of syntactic constructions. Previous structural priming studies involving children with DLD were carried out primarily in a single experimental session, and hence the impact of the children’s immediate learning in a longer time frame was unclear. This study aimed at investigating the implicit learning and the long-term learning retention of the Cantonese indirect object relative clause in children with DLD and children with typical development (TD) using a structural priming task. Twenty Cantonese-speaking 7- to 9-year old children with DLD and 20 age-matched children with TD participated in a structural priming task which were given twice, 1 week apart. The children’s immediate learning of the target relative clause during the priming tasks was examined. The children’s long-term learning retention was also evaluated using five probe tests arranged across a 3-week span. Children’s attempts to produce the target relative clause in the tests was examined and compared between the two groups. Results of the study were analyzed using the generalized linear mixed effects model. Both the number of exposures and group membership had an effect on the production accuracy of the target relative clause. All children were more accurate in the production of the target when the number of exposures increased. The children with DLD demonstrated weaker immediate priming effects when compared with the children with TD. For long-term priming effects, the children with DLD demonstrated immediate learning retention after the prime task and also long-term retention 1 week after the prime task. For the long-term retention, the two groups of children performed similarly. The age of the children was also found to have an effect on the children’s long-term learning retention. Similar error patterns were noted among the two groups’ attempts to produce the target relative clause. Findings in this study support the view that children with DLD demonstrate deficits in implicit learning. A multi-factorial explanation of children’s problems with grammatical constructions in relation to their implicit learning deficits is discussed. These factors may include weaknesses in lexical semantics, knowledge in thematic role, attention control, inhibitory control, and working memory. It is suggested that intervention should focus on two things. One is to support children with DLD’s knowledge in lexical semantics and thematic roles that are relevant to the learning of the target through visual aids. Two is to manipulate the salience and dose number of the exemplars of the target construction.