Neurocognitive risk factors for developmental dyslexia

• ¹ University of Jyväskylä, Department of Psychology, Finland
• ² University of Oslo, Department of Education, Norway

Purpose: The role of various risk factors for dyslexia, a neurodevelopmental reading disorder, has been debated for several decades. Phonological processing problems are widely acknowledged as being linked to dyslexia hampering learning of grapheme-phoneme correspondences. However, relatively little is known of neurocognitive mechanisms underlying phonology, reading related cognitive skills and their interaction with later reading problems. As dyslexia is known to be highly heritable, by studying brain responses of infants of dyslexic parents, it is possible to follow up infants who will encounter reading difficulties later in school-age. During early school years reading skills also develop at different rates. Therefore, examining the relationship between early brain responses and reading proficiency in early reading phase and later in adolescence is important and can reveal atypical neurocognitive development associated with dyslexia. As the digital media and Internet have changed literacy practices in many ways requiring new reading skills and strategies, such as locating and evaluating relevant information, we have also expanded our research on learning difficulties to the digital learning environments. Our aim is to increase understanding of internet reading and search skills and interconnections between internet reading, cognitive skills, and brain processes, to map the challenges which students with learning difficulties meet, and finally to promote pedagogical practices for assessing internet reading skills. Methods: We review findings from the Jyväskylä Longitudinal Study of Dyslexia (JLD), in which we have investigated whether dyslexic children with familial risk background would show atypical auditory/speech brain activation as measured by brain event-related potentials (ERPs) already at birth, at six (6) months and at the third (3) grade at 9-10 years of age. We examined associations of the infant brain responses to later pre-reading cognitive skills and literacy outcome at the 2nd (at 8-9 years) and 8th grades (at 14 years). In terms of digital reading, we studied 11-13 year-old school-aged children at the sixth grade in multidisciplinary interconnected studies in eSeek-project (see the poster by Otieno et al., this conference). About 400 sixth-graders, including students with learning difficulties, participated in a large scale internet skill assessment. About 150 of them participated in the laboratory setting in individual cognitive tests, simulated internet task involving eye-tracking, and brain event-related potential measurements related to reading comprehension and sub-components of attention. Results: In the JLD study, the children diagnosed with dyslexia at school-age and who had familial risk background showed atypical auditory/speech processing for various sound features, including non-speech and speech sounds, already at infancy. Atypical brain activation also persisted until school-age, albeit with different response patterns. Infant brain responses also correlated to childhood language and pre-school age reading related skills, including phonological processing and letter-naming skills, as well as to reading and writing skills at school age. The correlations were, in general, more consistent among at-risk children. Developmental changes were observed in these associations. Importantly and somewhat surprisingly, the infant ERPs at 6 months predicted reading speed up to the 8th grade. Further, speech perception at behavioral level differed between dyslexic and typical readers in school age at the 3rd grade, but not in all dyslexic readers. Similar findings from the large scale longitudinal Dutch Dyslexia Programme (DDP) show that infant brain responses to speech sound changes are related to familial risk for dyslexia and also predict later reading skills at school-age. Our results from Internet reading in eSeek-project showed that dysfluent readers at the sixth grade meet extra challenges, especially when evaluating commercial web-sites. The eye-tracking study showed that children with dyslexia use poorer strategies in selecting internet search results. Conclusion: Overall, our findings suggest developmental differences in the organization of the neural networks sub-serving auditory/ speech perception, with cascading effects on later reading related skills and reading fluency. However, the evidence also suggests that atypical basic processing skills alone are not likely a sufficient reason for dyslexia, but rather one endophenotype /risk factor supporting the multi-risk factor model of learning difficulties. Challenges remain, therefore, for the individual identification of high risk children. Further, our multidisciplinary approach to the Internet reading show that children with reading difficulties are at an elevated risk in learning 21st century digital skills. Further research is needed to increase scientific knowledge on digital reading in order to develop assessment tools and guidelines for instruction and teaching of internet readin