In this study, we are focusing on the crucially important transition from “learning to read” to “reading to learn” that children go through in the early grades. Our specific objective is to characterize how individual differences in reading sub-skills relate to children’s patterns of brain activation, and how these relationships change during the most important transition in reading development. Theories such as the lexical quality hypothesis (Perfetti, 2002) and the neural noise hypothesis (Hancock et al., 2017) posit that better readers have more precise and differentiated representations of the sounds, spellings, and meanings of individual words than poor readers. Our study will test this hypothesis using functional magnetic resonance imaging (fMRI). While data from behavioural studies has been invaluable in developing and testing these theories, it is limited because the measures depend on children’s overt judgements and behaviours; such data does not tell us how children’s brains are actually representing the information – and these representations are what form the basis of children’s judgements. Brain imaging offers the possibility to actually see how information about individual words is represented in children’s brains—and how distinct these representations are for different words.
In our study, children in grades 2 and 3 will complete a battery of standardized tests of reading and related skills, as well as a series of fMRI scans in which they will read single words. The fMRI data will be analyzed using representational similarity analysis (RSA), an approach that has been used successfully in the past to study the fine-grained brain activity patterns characteristic of individual stimuli such as pictures or words (Kriegestkorte, 2008; Fischer-Baum et al., 2017). Critically, RSA allows us to determine whether the degree of similarity (or dissimilarity) between words, in terms of dimensions like phonology, orthography, and semantics, correlates with the (dis)similarity of spatially -distributed patterns of brain activity evoked by those words. That is to say, if two words have similar spellings, then their pattern of brain activity should be more similar to each other than a pair of words with dissimilar spellings. Furthermore, following the lexical quality hypothesis, we predict that children who are better readers will show more distinctive patterns of fMRI responses to different words than will poorer readers.
Leveraging cutting-edge tools from cognitive neuroscience can inform educational theory, and get to the heart of why some children struggle to read. This understanding can in turn lead to more sensitive and precise assessments of reading, and approaches to teaching and remediating reading skills.
Fischer-Baum, S., BruggFemann, D., Gallego, I. F., Li, D. S. P. & Tamez, E. R. Decoding levels of representation in reading: A representational similarity approach. Cortex 90, 88–102 (2017).
Hancock, R., Pugh, K. R., & Hoeft, F. (2017). Neural Noise Hypothesis of Developmental Dyslexia. Trends in Cognitive Sciences, 21(6), 434–448.
Kriegeskorte, N., Mur, M. & Bandettini, P. A. Representational similarity analysis – connecting the branches of systems neuroscience. Front. Syst. Neurosci. 2, 1-28 (2008).
Perfetti, C. A., & Hart, L. (2002). The lexical quality hypothesis (L. Verhoeven, C Elbro, & P. Reitsma, Eds.; Vol. 11, p. 189 213).
Similar Projects
