During parafoveal processing, skilled readers encode letter identity independently of letter
position (Johnson, Perea, & Rayner, 2007). If dyslexia is linked to deficits in attention
allocation and letter position encoding (e.g. Whitney & Cornelissen, 2005), letter identity and
letter position information may not be extracted from the parafovea for dyslexic readers. In
two experiments, eye movements of skilled readers and readers with dyslexia were recorded
during a boundary paradigm experiment (Rayner, 1975). The participants in Experiment 1
were adults, whereas those in Experiment 2 were children. In both Experiments,
parafoveal previews were either identical (e.g. nearly), a transposed-letter preview (e.g.
enarly), or a substituted-letter preview (e.g. acarly). Dyslexic and non-dyslexic readers
demonstrated orthographic parafoveal preview benefits during silent sentence reading and all
reading groups encoded letter identity and letter position information parafoveally. However,
dyslexic children showed a delayed preview benefit compared to typically developing
children matched on chronological age, but similar results to typically developing children
matched on reading age. Dyslexic adults, however, showed a greater dependence on letter
position information for lexical identification during parafoveal processing compared to the
skilled adult readers. Furthermore, whilst dyslexic readers showed the usual difference in
foveal eye movement behaviour compared to age matched controls (e.g. longer viewing
durations, more fixations), dyslexic children also showed different eye movement patterns
compared to typically developing children matched on reading age. Thus, the current findings
demonstrated that dyslexic readers show consistent and dyslexic-specific reading difficulties
in both foveal and parafoveal processing during silent sentence reading.