Neurolinguistic profiles of developmental dyslexia in a longitudinal perspective
03 / 2011 - 02 / 2016
Aim of the proposal is to obtain a deeper understanding of the underlying phonological deficits in developmental dyslexia. Dyslexia is a multi-faceted, heterogeneous condition, which can be studied at different levels of aggregation, including genetic, neurological, cognitive, linguistic, behavioral, and environmental. According to current theories, it is assumed that dyslexia is caused by a phonological core deficit of genetic origin. Much progress has been made in understanding the early neurolinguistic variables (the past state) that contribute to the manifestation of dyslexia (the present state). In particular, it has been shown that dyslexia has a strong familial/genetic etiology, and that neurolinguistic precursors during infancy and early childhood precede the manifestation of dyslexia. In addition, in the perspective of further development (the future state), it has become clear that people with dyslexia do develop reading skills, suggesting patterns of both neurolinguistic ?normalization? and ?compensation?. However, two questions are unanswered. First, although about 40% to 60% of children from dyslexic parents are diagnosed with dyslexia (from now on labeled DYS), the remaining at-risk children learn to read and spell within normal ranges (nonDYS). It is unclear how the differences between affected and unaffected children in reading abilities relate to neurolinguistic ?normalization? or ?compensation? patterns at the level of automatized learning. Normalization would most likely occur in cases with low severity of involvement of the underlying phonological deficit, such that phonological representations and grapheme to phoneme connections do develop, possibly in interaction. Compensation implies overcoming the effects of the core phonological deficit by other mechanisms for sound ? letter matching, such as making use of lexical access, and would most likely occur in more severe cases of phonological deficit. A related question is whether processes of normalization or compensation in the present state at the age of 11-12 years (5th-6th grade) can be predicted on the basis of the past state of neurolinguistic functioning at the age from 0-9 years, and, in turn, predict the future state when the participants are at the secondary school (age 13-14 years). Second, there is large variability in the developmental trajectory of reading acquisition and the ultimately reached level of reading proficiency among children diagnosed with dyslexia. The typical reading acquisition process is a dynamic process in which reading strategies emerge and neural language networks for the reading task evolve. Whereas phonological processes localized at superior-dorsal temporal regions are crucial at the initial reading stage, in subsequent stages whole-word reading strategies, taking a more ventral route, and semantic integration processes for text comprehension, localized at inferior frontal sites, are developed. The question is to what extent and under what conditions the underlying phonological deficit in dyslexic readers hampers the emergence of these higher-level language networks. Are the large differences in ultimately reached functional reading ability among genetically at-risk readers caused by a quantitative difference, that is slower development of the basic reading process in poor readers, or qualitative differences, that is are there differences in strategy by the development of higher order compensating mechanisms. A related question is whether the differences in the present state at the age of 11-12 years predict the future state when the participants are at the secondary school (age 13-14 years). To investigate these questions, the present proposal puts dyslexia in a longitudinal neurolinguistic perspective. Using the well-documented longitudinal Dutch Dyslexia Programme (DDP), in which children were followed between the ages of 0 and 9 years, hypotheses about the present state, and the relation with the past and the future state are put to test. Project 1 makes use of a model of distributional learning to study automatised normalization and compensation processes in the present state at the age of 11-12 years. To relate the present state to the past state, the project makes use of neurolinguistic DDP-data to trace back the developmental parameters leading to DYS and nonDYS, as compared to controls without genetic risk (C). Following the participants, the relation with the future state at age 13-14 years is also investigated. Project 2 monitors the higher order reading acquisition process by means of fMRI combined with EEG (ERP) measurements while the children perform increasingly complex reading tasks requiring integration of phonological, lexical and semantic processes, at present and future state. The longitudinal approach of DDP in combination with the present programme, spanning an age range from 2 months to 14 years, offers the unique opportunity to find answers to fundamental questions about reading acquisition and dyslexia.