Despite having mastered basic word reading skills, large numbers of readers struggle to understand what they read. More worrisome still is the over-representation of minority and low-income populations in the ranks of struggling readers, which is linked to poor health outcomes. At the same time, there is very little agreement among psycholinguists and education researchers as to what factors drive variability in language understanding. This research program tests two factors that have been argued to explain variability in grammatical processing and language comprehension abilities. The first are differences in cognitive abilities (i.e. working memory, processing speed, and phonological ability). The second are differences in experience with language. We test these two possibilities in a series of experiments examining individual differences in language comprehension in young readers and adults. We have 2 specific aims: 1) To determine whether syntactic processing is guided by cognitive constraints or language experience; 2) to determine whether correlations between language experience and comprehension skills that were discovered in pilot work are the result of previous experience with diverse grammatical structures or the result of improved parsing efficiency derived from practice. The outcomes of the proposed experiments have implications for core questions in language research. Evidence that working memory or other cognitive factors predict language skill is most consistent with theories that domain general aspects of cognition are engaged in understanding language. Evidence that language experience predicts variability in reading skill is more consistent with theories that posit that statistical representations drive language processing.
This work represents a novel exploration of individual differences in reading ability in both children and adults, with the goal of understanding which factors lead to poor comprehension. Low literacy attainment is linked with poor health outcomes and increased levels of infant and child mortality in the population (DeWalt & Pignone, 2005), and is even more closely associated with poor health than the number of years of school completed (Baker et al., 1996).
