Predicting academic success in students with dyslexia: a multimodal SES, cognitive and neural classifier approach – DYSuccess

According to PISA 2015, approximately 20% of 15-year-old French students do not attain the baseline level of proficiency in literacy. Students with dyslexia, who manifest severe and persistent poor word decoding, low reading fluency, and poor spelling skills, are more likely than typical students not to attain this level. Nevertheless, some dyslexics successfully manage to study at the university level. The general aims of this project are 1) to investigate the socio-demographic (SES), cognitive and neural correlates of risk and protective factors in French high-school students (aged 15-17) with dyslexia (Work-Package 1, WP2, WP3, & WP4); and 2) to propose and validate a neurocognitive model that will predict academic success of high-school students with dyslexia by classifying well-compensated versus less-compensated, and those at risk for school failure or dropout, using a multimodal socio-demographic, cognitive and neural machine learning classifier (WP5). This work will help to inform the development of policies and targeted intervention schemes to fight school failure or dropout, thereby addressing inequality in academic success. Ultimately, this may lead to early identification of potential learning disorders and personalized education including individual learning profiles.