Eye Movements and Phonological Parafoveal Preview: Effects of Reading Skill

Abstract

Eye moves of skilled and less skilled readers were monitored as they read opinions containing a target word. The boundary paradigm was used similar that when their eyes gibbeted an invisible boundary location, a preview word changed to the target word. The preview could either be identical to the target word (beach as a preview for beach), a homophone of the target word (beech as a preview for beach), an orthographic command (bench as a preview for beach), or an unrelated consonant string (jfzrp as a preview for beach). Consistent with prior research, skilled readers obtained more preview benefit from the homophone preview than from the orthographic preview. The les skilled readers, however, did not exhibit such an effect. The issues indicate that less skilled readers do not use phonological collection of lawss to integrate information across organ of vision movements. Indeed, the results also indicate that les skilled readers do not display normal preview benefit effects.

Readers obtain information not single from the word they are looking at, on the contrary also from the word to the right of fixation in the parafovea (Rayner, 1998) Pollatsek, Lesch Morris, and Rayner (1992; diocese also Henderson, Dixon, Petersen, Twilley, & Ferreira, 1995) were the first to demonstrate that readers use phonological collection of lawss to integrate information across successive saccades. In this article, we examine the expanse to which reading skill influences phonological parafoveal preview effects

Pollatsek et al. (1992) examined the result of phonological preview information by the agency of embedding target homophones in determinations and then varying the parafoveal preview that readers received (see Figure 1 for an illustration of by what means parafoveal preview was manipulated). Four preview conditions were used: in the first, the target and the preview were identical (beach was the preview for beach); in the next to the first the preview was phonologically the same as the target (beech - beach); in the third, the preview was an orthographic superintendence word (bench - beach); and in the fourth, a word that was orthographically and phonologically actual dissimilar to the target serv as the preview (house - beach). The target word was read faster in the homophone preview condition compared to the orthographically similar preview condition. This hints that phonology is aiding word identification because the homophone and orthographic dominion government word share the same number of overlapping alphabetic characters with the target. Pollatsek et al. also obtained preview benefit for homophone that were not orthographically similar to each other (shoot as a preview for chute and vice versa).

More lately Pollatsek et al.'s (1992) flows were replicated by Miellet and Sparrow (2004) for French readers and by means of Tsai, Lee, Tzeng, Hung, and Yen (2004) for Chinese readers. the couple of these studies used the boundary paradigm (Rayner, 1975) used through Pollatsek et al. Also, in a variation of the boundary paradigm used in reading research, Pollatsek, Tan, and Rayner (2000) obtained evidence that Chinese readers rely upon phonological codes when integrating information across saccades.

While this research demonstrates that phonological previews aid in the processing of the target word, evidence from individual difference studies give an inkling ofs that phonology is not used in the same manner through all readers. Very little research has been leadershiped on individual differences in phonological processing among adult skilled readers. However, a certain quantity of research suggests that differences in skill horizontal even among college students, can affect phonological processing (Jared, raise & Rayner, 1999; Unsworth & Pexman, 2003) Jared et al. (1999) demonstrated that les skilled readers (skill horizontal was determined by performance upon the Nelson-Denny reading test) were les likely than more skilled readers to notice a homophone error than an orthographically matched error in a proofreading task. Unsworth and Pexman (2003) base that less-skilled readers (skill horizontal was again determined by the Nelson-Denny Reading Test) made more rejoinder errors in a lexical decision task compared to more skilled readers for pseudo-homophones than nonhomophone ascendencys They postulated that the discrepancy between more and les skilled readers hints less skilled readers have "les efficient orthographic-phonological mappings" than more skilled readers (Unsworth & Pexman, p 72)

In the near experiment, we used the boundary paradigm to replicate previous work upon phonological preview effects (Miellet & Sparrow, 2004; Pollatsek et al., 1992) However, reading skill horizontal of the participants in the experiment was obtained to determine if less-skilled readers utilize phonological collection of lawss to integrate information across saccades. In line with the previously mentioned studies upon individual differences, we expected that less-skilled readers might benefit les from phonological preview information than more skilled readers because more of their resources may be devot to processing the general word, leaving less capacity available to effectively proces parafoveal information.