Van der Haegen, L. & Brysbaert, M.
Ghent University
The split fovea theory (SFT) states that centrally and parafoveally presented information is organized in the same way: Letters at the left/right side of fixation are initially sent to the right/left hemisphere respectively. As a consequence, the beginning of a word is the optimal viewing position for readers with left hemispheric language dominance because this makes most letters of a word fall into the right visual field. Until now, the assumptions of SFT were never directly tested under strict methodological settings such as stimulus sizes within the central 3° of vision and strict fixation control. In the present study, subjects with typical left and atypical right language dominance named three-, four- and six-letter words at different fixation locations. An eye-tracking device controlled the fixation position of both eyes and registered naming latencies for all possible letter position fixations. Results showed that left dominant subjects were fastest when fixating at the word beginning whereas the optimal viewing position of right dominant subjects was situated more towards the end of the word. Consequently, visual word recognition models should take into account that interhemispheric transfer is needed for both parafoveal and foveal word recognition.