Is the role of set for variability in word reading influenced by conditions leading to partial decoding?

Purpose: Set for variability (SfV; the ability to disambiguate the mismatch between the decodedform of a word and its correct pronunciation) has been shown to predict regular and irregular word reading in children across both transparent and opaque orthographies. It has been offered that SfV is a crucial part of the decoding process associated with orthographic learning. Consistent with this hypothesis, it has been suggested that SfV is more important when partial decoding occurs such as in irregular words or when children lack the decoding skills to fully decode the word. The purpose of this study was to investigate whether the influence of SfV skill on the prediction of word reading is impacted by a word’s regularity and a child’s decoding skill. Methods: The present study examined whether the relation between SfV and word reading is moderated by a word’s spelling-to-pronunciation transparency rating and a child’s decoding skill using a cross-classified random effects model of 491 children and 160 words. Results: SfV was shown to be more important under conditions that may lead to a partial decoding such as when decoding skill is low or word transparency is low and was most important when both decoding skill is low and word transparency are low. Conclusions: This suggests that when children with poor decoding skills read irregular words whose pronunciation is not readily available from its spelling, SfV is a large contributor to whether that word will be read correctly.