Within the framework of my dissertation project I analyzed the formal linguistic as well as the socio-pragmatic skills of preschool children in different play situations. The aim of my research was to find out the abilities and deficits of specific language impaired (SLI) children in child-to-child communication and especially in role-play compared with their age-matched peers.
Most of the children with SLI have limited socio-pragmatic skills (Bishop, 2000; Craig, 1993; Fujiki et al., 2001; Liiva & Cleve, 2005; Marton et al., 2005). In summary, they exhibit difficulties in social pragmatics including negotiation, conflict resolution, and interaction with peers. Nevertheless, there is little consideration of communicative skills in the diagnosis and therapy of German speaking SLI children.
Yet, child-to-child interaction is very important for the development of communicative and social skills (Andresen, 2002). This assumption relies on Vygotsky's thesis that in role play preschool children are acting in the zone of proximal development, although their ability to cooperate with other children is only at a nascent stage (Andresen, 2005). Andresen specifies this thesis with respect to language development and is able to show with her empirical investigations that language is the central means of creating pretence.
This way children seem to enhance their socio-pragmatic linguistic skills effortlessly. There are still insufficient studies, examining the child-to-child interaction of German speaking children with SLI.
So the main questions for my study are:
Which differences can be found between dyadic interaction of children with SLI and normally developed children?
Are there differences in terms of linguistic skills in SLI children in situations with different restrictions?
Is there a correlation between socio-pragmatic and formal-linguistic skills in SLI children?
The participants of the study are six preschool children with SLI and six normally developed children as controls (aged 4-6 years). The study took place in familiar environment of the children. They were observed in naturalistic interactions on two occasions. The first situation was a relatively free play situation and the second one was more structured.
The outcome of the study, on the one hand, an overall group comparison between the SLI and the control group showing SLI children produced fewer grammatically complete clauses, fewer metacommunicational utterances but more grammatical errors. Although highlighting group differences in each single situation, no significant group differences were found for grammatical complexity and metacommunicative expressions but only significant differences for the number of errors in the free play situation. By comparing the linguistic competences in the different situations for each of the groups, both SLI children and normally developed children showed longer clauses and more grammatical errors in the free play situation. It was remarkable that, unlike the control group, the SLI children produced fewer metacommunicational utterances in the free play situation than in the structured play situation. Furthermore correlations were found between grammatical complexity and metacommunicative expressions in both groups. So, in summary the data support the assumption that linguistic competence is affected by the kind of situation for all children. It seems that especially SLI children benefit from situations with fewer organizational restrictions. Moreover, there seems to be an overall correlation between the number of metacommunicative expressions and the grammatical complexity. Interestingly enough in my study the data did not point out evidence for general communicative problems in SLI children.
However, when interpreting the data it has to be kept in mind that the outcome of the study is based completely on quantitative data. The next step would be to look at the data with qualitative methods in order to expose possible variations in communicative strategies of SLI as well as normally developed children.