Abstract
Background: The production of causal sentences (e.g. The house went on fire because the girl was playing with matches) is a key component of the ability to produce explanations, which in turn is an important aspect of children's developing discourse skills. While informal observations by professionals suggest that children with language impairments often have difficulties in producing (and comprehending) causal sentences, there is a dearth of systematic research evidence on the scope and characteristics of such difficulties.
Aims: The study reported herein aimed to establish the scope and characteristics of the difficulties that children with language impairments have with causal sentence production. In particular, it investigated whether they have difficulties with (1) producing causal connectives (because and so), (2) producing semantically appropriate causal sentences, and (3) coordinating the production of more than one clause.
Methods & Procedures: The performance of 5–7‐year‐old children with language impairments (the LI group, n = 30) was compared with that of typically developing chronological age peers (the CA group, n = 30) on four elicited production tasks designed to encourage children to talk about pictures of causally related events. The tasks required the children to answer causal questions, to complete and imitate causal sentences, and to produce full causal sentences.
Outcomes & Results: Although the groups did not differ in the overall frequency with which they used causal connectives, the LI group produced fewer causal connectives than the CA group on tasks involving higher processing demands, as well as producing a lower percentage of semantically appropriate responses on most tasks. The LI group found imitating causal sentences more difficult than answering causal questions and completing causal sentences, whereas the CA group showed a similar level of performance across these three tasks.
Conclusions: Although 5–7‐year‐old children with language impairments have causal expressions in their repertoire, they have marked and extensive problems in using these expressions appropriately and in producing full causal sentences even when these are modelled for them. Given the educational importance of explanation, there is a clear need for interventions to target both the semantic appropriateness of causal sentence production and the frequency of use of causal connectives, especially in contexts where children have to cope with the additional processing demands associated with producing two‐clause sentences and/or with producing causal sentences autonomously.
Notes
1. There is also a third type of causal sentence structure in which because occurs in sentence‐initial position (i.e. because–cause–effect, as in Because the girl was playing with matches, the house went on fire). Thus, the choice of connective depends on its position as well as on the order of the clauses. However, this third type of sentence is likely to be rare in children's speech and so will not be considered further here.
2. It should be noted that the comparisons between autonomous and prompted production were carried out only for the tasks involving the production of two clauses. The Asking Questions task could perhaps be regarded as an autonomous one‐clause task (although classifying it as ‘one clause’ is debatable for what next questions), but it also differs from the prompted one‐clause tasks (Answering Questions and Completion) in other respects which could make comparisons between these tasks difficult to interpret.