The Underlying Dimensionality of Spoken Word Recognition: Continuous development through the school-age years

Spoken word recognition relies on rapid, real-time competition among lexical candidates, yet it remains unclear which aspects of this competition change across later childhood and how they relate to individual differences in language and reading development. We tracked 263 children for three years (starting in grades 1-3) and measured spoken word recognition annually using the Visual World Paradigm alongside standardized language and reading assessments. Principal component analysis of fixation dynamics identified five orthogonal dimensions explaining 80% of the variance, revealing a multidimensional architecture. Growth-curve analyses showed that the second dimension—activation rate—showed robust developmental change and uniquely covaried with language ability, independent of non-linguistic processing and cognitive regulation. A third dimension—competitor suppression—was also linked to language ability, particularly among children with weaker language skills. Together, these findings demonstrate continued reorganization of spoken word recognition across later childhood, with distinct competition mechanisms differentially linked to language development.