The relationship between language skills, social cognition and externalising behaviour in Primary school aged boys

There is evidence of a high incidence of language difficulties (LD) amongst boys with externalising behaviour (EB); however we still have little understanding about why they co-occur. This 3 part study investigates aspects of this relationship framed within a biopsychosocial model and with a focus on pragmatic language skills and social cognition. Firstly, this study seeks to replicate recent research that has indicated a close association between pragmatic language skills and EB. It is the first study to consider the strength of this association while controlling for other variables known to commonly co-occur with LD and EB: aspects of the child's ability (word decoding and nonverbal cognitive skills), and aspect of their environment (parenting stress, maternal education and family set up). Secondly, this study furthers our understanding of the social cognitive and friendship skills of boys with LD, in both areas through investigating reasons for variance in ability. A better understanding of an LD population feeds into our understanding of EB due to the high proportion of boys with EB who have coexisting LD. Thirdly, this study measures social cognition of boys EB while accounting for the role of LD, thus investigating whether difficulty with these tests is associated with the high rate of LD in this population. Previous studies have not adequately considered this. Method: Boys aged 8 to 11 years receiving additional support in school were assessed for LD and EB and two groups (not mutually exclusive) were identified: boys with LD (n=31) and boys with EB (n=35). A control group of typically developing boys matched for age and SES were also identified (n=42). For part one, participants completed assessments of language skills, word decoding and non-verbal cognitive ability. Teachers completed a checklist to provide a measure of pragmatic language skills. Parents completed questionnaires to provide measures of parenting stress, family set-up and maternal education. For the second and third part of the study participants' social cognition was assessed and parent and teachers completed a checklist for measurement of emotional and behavioural difficulties including friendship skills. Results and discussion: In the first part of this study, all variables measured were found to be significantly associated with EB, as would be expected within a biopsychosocial model in which many factors interact with each other in the development of EB. However, particularly close associations were found between pragmatic language skills and EB (replicating previous research in this area), followed by language skills. This indicates a close and specific association between communication skills and EB, even when other closely associated variables are accounted for. In part two, amongst boys with LD, different social cognition assessments were found to be associated with different aspects of communication skills. One was most closely associated with structural (particularly expressive) language and the other two with pragmatic language. This indicates that they are tapping different constructs and highlights the difficulty using a proxy to give an indicator of social cognition; it is not easy to assess. Variability in friendship skills of boys with LD was not associated with any aspect of communication skills. Significant correlations were found with two of the social cognition test scores only. This is at variance with research indicating a link between friendship and receptive language skills. In part 3, the EB group scored significantly lower than the Control group in tests of social cognition, however EB was not found to be linked with social cognition score. Only boys with LD (with or without EB) scored significantly lower than the Control group. This has implications for previous research into the social cognition of boys with EB which has not fully considered the high proportion with LD and the extent this may be contributing to their low social cognition scores. Implications for practice and future research are discussed.sub_shsunpub662_ethesesunpu