Eye movements were monitored in 4 experiments that explored the role of parafoveal word length in reading. The experiments employed a type of compound word where the deletion of a letter results in 2 short words (e.g., backhand, back and). The boundary technique (K. Rayner, 1975) was employed to manipulate word length information in the parafovea. Accuracy of the parafoveal word length preview significantly affected landing positions and fixation durations. This disruption was larger for 2-word targets, but the results demonstrated that this interaction was not due to the morphological status of the target words. Manipulation of sentence context also demonstrated that parafoveal word length information can be used in combination with sentence context to narrow down lexical candidates. The 4 experiments converge in demonstrating that an important role of parafoveal word length information is to direct the eyes to the center of the parafoveal word.
Keywords eye movements; reading; parafoveal length informationIn modern alphabetic writing systems, blank spaces segment strings of letters into meaningful units (words) that contribute to the efficiency of reading. When spaces are deleted or masked, reading is significantly slowed (Epelboim, Booth, & Steinman, 1994;Fisher, 1976;Malt & Seamon, 1978;McConkie & Rayner, 1975;Morris, Rayner, & Pollatsek, 1990;Rayner, Fischer, & Pollatsek, 1998;Spragins, Lefton, & Fisher, 1976). There are three reasons why interword spaces facilitate reading. First, blank spaces reduce lateral masking of the beginning and ending letters of words (Bouma, 1973) making them easier to see. Second, interword spaces specify the length of upcoming words and provide effective targets for eye movements (Rayner, 1979). Third, word length may be used to constrain the number of possible word candidates. The present article explores the relative importance of these different factors.Correspondence concerning this article should be addressed to Barbara J. Juhasz, Department of Psychology, 207 High Street, Wesleyan University, Middletown, CT 06424. E-mail: bjuhasz@wesleyan.edu. There are, of course, writing systems, such as Chinese and Thai, which do not contain interword spacing in which readers still read effectively. However, research on Thai (Kohsom & Gobet, 1997) indicated that inserting spaces between words actually increases the reading rate in that language, and research on Chinese (Bai, Yan, Liversedge, Zang, & Rayner, in press) indicated that inserting spaces between characters interfered with reading, whereas inserting spaces between words did not (see also Sainio, Hyönä, Bingushi, & Bertram, 2007, for research on spacing in Japanese). Also, work with German three-lexeme compound words has shown that inserting spaces between the lexemes actually reduces overall reading time on the compounds (Inhoff, Radach, & Heller, 2000), even though it is orthographically illegal to do so.
NIH Public AccessTo better understand why interword spaces are beneficial to reading, it is important to consider...