Robin van Rijthoven

129 Compensatory role of verbal learning and consolidation in reading and spelling 5 the two groups were rather small and therefore somewhat unstable. Thirdly, different learning strategies that were taught to children in their schools could explain the contradicting results. Our first research question was whether verbal learning and consolidation predict reading and spelling in children with dyslexia compared to controls. We expected verbal learning and consolidation to be associated with reading and spelling outcomes, even more so in childrenwith dyslexia (see Shaywitz et al., 2003). We found that verbal learning was not related to word reading, negatively related to pseudoword reading for children with dyslexia (this negative relation disappeared when controlling for phonological awareness), and positively related to spelling for both childrenwith dyslexia and typically developing children. The effects were small to medium. Furthermore, no compensatory role of verbal consolidation was found. The finding that verbal learning was negatively related to pseudoword reading could indicate that especially children with decoding problems train and use their verbal learning ability in an attempt to overcome these problems. Pseudoword reading is in the end a task that requires decoding skills mostly. The fact that this negative effect disappeared when we added phonological awareness to the model indicates that these decoding difficulties are phonologically based. The fact that verbal learning did not positively influence children’s reading contrasts findings by Tijms (2004) but the positive influence of verbal learning on spelling levels among children with dyslexia is in line with findings by Tijms (2004). In addition, we showed the unique influence of verbal learning on spelling in both children with dyslexia and typically developing peers. Overall, the results fit the idea that better verbal learningmay support the specificity and redundancy of the phonological lexicon and therefore stimulates literacy development (Shaywitz et al., 2003), but shows that the effect is limited to spelling only. This can be explained by the fact that spelling relies more on phonological representations compared to reading (Landerl & Wimmer, 2008). This outcome seems to be in line with the statistical-learning perspective on spelling development, which emphasises that the spelling of children reflects the input to which children have been exposed. This input is filtered through their learning mechanisms (Pollo et al., 2007). Children with stronger verbal learning abilities (their learning mechanism) can learn more from exposure to verbal input, which benefits their spelling. The results showed no compensatory role of verbal consolidation. This is in line with the conclusion of Tijms (2004) who concluded that once a deeper semantic level is activated, no further problems appear to be encountered and consolidation of verbal information can take place as normal. It could also be due to the fact that verbal consolidation is closely related to automaticity (Manoach & Stickgold, 2009). Indeed, in