Robin van Rijthoven

144 Chapter 6 after an intervention among children with dyslexia. Individual differences in strengths and weaknesses in phonological abilities, verbal learning, and verbal consolidation did not influence the progress during the phonics through spelling intervention but semantic abilities did (at least for spelling). This highlights the importance of this domain in the triangular framework. With regard to our second research question, we can conclude that stimulating orthographic representations with a phonics through spelling intervention benefits reading and spelling development of children with dyslexia. Semantic abilities could be seen as a compensatory factor to optimize the response to intervention outcomes for spelling. Compensating reading and spelling in children with dyslexia revisited The findings in this dissertation indicate that relative strengths and weaknesses in semantic abilities and stimulating orthographic representations influence the reading and spelling development of children with dyslexia. Results of our studies revealed that semantic abilities and the stimulation of orthographic representations in their own way could be seen as compensatory factors for the poorly developed phonological pathway of children with dyslexia (Conrad, 2008; Nation & Snowling, 2004). The two compensatory factors under consideration were derived from the triangular framework (Seidenberg & McClelland, 1989). This framework consists of phonological, orthographic, and semantic representations. The bi-directional relationship between these factors forms an efficient and recurrent network that is necessary to efficiently read and spell. In alphabetical languages the bi-directional relation between phonological and orthographic representations seems to be the most prominent factor facilitating proficient reading and spelling (see Bosman & Van Orden, 1997). However, children with dyslexia have underspecified and inaccurate phonological and orthographic representations (Conrad, 2008), mostly due to the underlying phonological deficit (Melby-Lervåg et al., 2012) and thereby develop reading and spelling problems. These children may need to activate compensatory pathways to learn to read and spell (Hoeft et al., 2011; Kearns et al., 2019), as is also indicated by brain studies that showed that different parts of the brain were activated during reading among children with dyslexia (Paulesu et al., 2001; Shaywitz et al., 2003). Our first hypothesis was that well-developed semantic representations could facilitate such an alternative route. Our results indicate that children with dyslexia are able to compensate for their phonological shortcomings by using well-developed semantic representations to facilitate reading indirectly via phonological awareness and rapid automatized naming and by facilitating spelling directly. We also found that verbal learning, which can be considered a more dynamic approach towards semantic abilities, was related to spelling as well, indicating that not only the semantic knowledge present is of importance but also the ability to learn new semantic information. Children with dyslexia may thus use their semantic abilities