Farmer, T. 1 , Engelhardt, P. . 2 , Myachykov, A. 3 & Greer, J. 3
1 University of Iowa
2 University of East Anglia
3 Northumbria University
Readers use knowledge about language to generate predictions for multiple aspects of upcoming words, and recent work suggests that predictions can be for surprisingly low-level form-based properties of the unfolding input. Phonological Typicality (PT) refers to the degree to which the sound properties of an individual word are typical of other words in its lexical category. The information captured by this measure is inherently statistical, implicating statistical learning as a key mechanism of acquisition of knowledge about these form-category mappings. When sentential context sets up a strong expectation for a noun to occur, the target noun elicits shorter fixation durations when its form-based properties are category typical, as opposed to atypical, of other nouns. This effect is absent, however, for readers with a dyslexia diagnosis. In the work presented here, we provide an overview of data from an individual differences experiment involving disordered and non-disordered readers in order to address why this effect is absent for those with a dyslexia diagnosis. I will focus most centrally on individual differences in statistical learning ability, along with interactions between statistical learning ability and perceptual processes, as key predictors of the strength of the typicality effect in either group.