Investigating the development of a developmental disorder : mapping the trajectory of lexical development in specific language impairment

Abstract

There is increasing consensus that to understand developmental disorders we must apply developmental theoretical models and methodologies. To develop a fully specified developmental model of a developmental disorder we must understand both the nature of the innate causal processing deficits of the disorder and also how these deficits in early processing mechanisms then change the developmental process. This study aimed to examine the second of these issues with respect to Specific Language Impairment (SLI) and so describe the altered trajectory of development in this group of children. Explanatory models which propose hypothetical trajectories of development from impaired processing mechanisms in the infant to the patterns of linguistic impairments typically found in SLI are beginning to be developed. To date however there is very little empirical research which maps these trajectories. This study sought to contribute to that necessary empirical data and so to our understanding of the development of SLI. In addition it aimed to consider whether the application of a developmental methodology and perspective adds to our understanding of this disorder. A series of longitudinal case studies of children with SLI were completed. The participants were seen for four blocks of comprehensive assessment of language processing, language knowledge and "language relevant" processing over a 15 month time period. Cross sectional data from 38 typically developing (TD) children was also collected for comparison purposes. The data presented represents a part of this larger study and focuses on the development of the lexicon in SLI. Lexical and phonological processing and their interaction with phonological working memory capacity are thought to be crucial to the ontogeny of SLI. A series of tasks were developed to create a window into the nature of the developing lexicon. Data is presented from a novel non-word repetition task which manipulated the phonological characteristics of the stimuli and from a fast-mapping task where both phonological and lexical variables were manipulated. The influence of these factors on performance and changes in their influence across development were examined. Analysis of the trajectory of development of the two measures in TD children showed evidence of increasing abstraction of sub-lexical/phonological knowledge from lexical knowledge across development. In addition the developmental trajectory of fast mapping abilities demonstrated a significant and radical shift in processing bias across the age range. This result suggests that functional reorganisation in the developing lexicon, and hence the speech processing mechanism, may be taking place and which may occur as a result of increasing sub-lexical/phonological abstraction. The developmental trajectories of the children with SLI suggest that this group of children develop a different lexical processing architecture from typically developing children which does not reach the levels of efficiency of TD children's speech processing mechanisms. There is tentative support for a deficit in schema abstraction across the lexicon and an absence of functional reorganisation. The possibility that these results represent entrenchment within a self-organising network, and the possible relationship to issues of timing and critical periods is discussed. In addition it appears that compensatory strategies for this inefficient speech processing architecture may result in impaired semantic learning and so may have effects on the wider trajectory of atypical language development in SLI. Applying a developmental emergent perspective to SLI and so considering trajectories of development rather than static group comparisons can begin to uncover the nature of change within an interactive system and the nature of interdependence of processing mechanisms across development. Such an approach holds promise for revealing the nature of SLI and providing a more ecologically valid explanation of this complex disorder. The implications of developmental emergent conceptualisations of language impairment for research methodologies, diagnosis and therapy are discussed.