Although previous research has shown that letter position information for the first letter of a parafoveal word is encoded less flexibly than internal word beginning letters (Johnson, Perea & Rayner, 2007;White et al., 2008), it is not clear how positional encoding operates over the initial trigram in English. This experiment explored the preprocessing of letter identity and position information of a parafoveal word's initial trigram by adults and children using the boundary paradigm during normal sentence reading. Seven previews were generated: Identity (captain); transposed letter and substituted letter nonwords in Positions 1 and 2 (acptain-imptain); 1 and 3 (pactain-gartain), and 2 and 3 (cpatain-cgotain). Results showed a transposed letter effect (TLE) in Position 13 for gaze duration in the pretarget word; and TLE in Positions 12 and 23 but not in Position 13 in the target word for both adults and children. These findings suggest that children, similar to adults, extract letter identity and position information flexibly using a spatial coding mechanism; supporting isolated word recognition models such as SOLAR (Davis, 1999(Davis, , 2010 and SERIOL (Whitney, 2001) models.Keywords: letter position encoding, parafoveal preview, eye movements, childrenThe purpose of this study was to examine how letter identity and position information are encoded during lexical identification in sentence reading by children and adults. Specifically, in this study, parafoveal preprocessing of letter identity and position information in a word's initial trigram by children and adults during silent sentence reading was explored.
Parafoveal Preprocessing in Children and AdultsResearch in parafoveal preprocessing in adults, using gazecontingent change paradigms (McConkie & Rayner, 1975;Rayner, 1975), has shown that readers not only process the fixated word but also extract some visual and linguistic information from the next word in the sentence, before it is directly fixated (see Schotter, Angele, & Rayner, 2012 for a review). Studies using the moving window paradigm have shown that in skilled readers, the effective visual field in reading (the perceptual span) extends over an asymmetrical area from 3-4 characters spaces to the left of the fixated word to 14 -15 character spaces to the right of fixation in alphabetic languages (McConkie & Rayner, 1975). Word identification occurs in the area closest to fixation (between 3 and 4 letters to the left and 6 or 7 letters to the right of fixation; Rayner & Bertera, 1979;Rayner, Inhoff, Morrison, Slowiaczek, & Bertera, 1981).With respect to the size of the effective visual field in reading for children, studies have shown that the perceptual span increases with age. Thus, 7-to 9-year-old children were found to have a perceptual span of 3 to 4 letter spaces to the left of fixation and 11 letters to the right; while the span was 3 to 4 letters spaces to the left and 14 letters to the right of fixation in 11-year-old children (Häikiö, Bertram, Hyönä & Niemi, 2009;Rayner, 1986;Sperlich, Sch...