Academic journal article Memory & Cognition

Eye Movements and Parafoveal Word Processing in Reading Chinese

Academic journal article Memory & Cognition

Eye Movements and Parafoveal Word Processing in Reading Chinese

Article excerpt

In two experiments, a parafoveal lexicality effect in the reading of Chinese (a script that does not physically mark word boundaries) was examined. Both experiments used the boundary paradigm (Rayner, 1975) and indicated that the lexical properties of parafoveal words influenced eye movements. In Experiment 1, the preview stimulus was either a real word or a pseudoword. Targets with word previews, even unrelated ones, were more likely to be skipped than were those with pseudowords. In Experiment 2, all of the preview stimuli had the same first character as the target. Target words with same-morpheme previews were fixated for less time than were those with pseudoword previews, suggesting that morphological processing may be involved in extracting information from parafoveal words in Chinese reading. Together, the two experiments dealing with how words are processed in Chinese may provide some constraints on current computational models of reading.

Reading is a complicated and dynamic mental process in which readers actively move their eyes to extract information for comprehension. During each fixation, decisions need to be made concerning when to move the eyes and where to direct the gaze. Meanwhile, not only the fixated word, but also the word(s) hi the parafovea are processed. Information extracted parafoveally facilitates processing on the subsequent fixation, which is referred to as the parafovealpreview benefit (Rayner, 1998). The range within which useful information can be extracted during a fixation is called the perceptual span (Rayner, 1998). For instance, the perceptual span extends 3 or 4 letters to the left and 14 or 15 letters to the right of the fixation point when English is read. In addition, it has been shown that factors such as word length, word frequency, and contextual predictability impact eye movement behavior (Calvo & Meseguer, 2002; Kliegl, Grabner, Rolfs, & Engbert, 2004; see Rayner, 1998, for a review).

Recently, computational models have been designed to account for the underlying mechanism (i.e., the coordination of visual, linguistic, and oculomotor systems) of eye movement control during reading (Engbert, Nuthmann, Richter, & Kliegl, 2005; Pollatsek, Reichle, & Rayner, 2006; Reilly & Radach, 2006; see Reichle, Rayner, & Pollatsek, 2003, for comparisons among models). Broadly speaking, these models differ in their assumptions about (1) whether cognitive processing is the driving force that triggers eye movements and (2) whether words within the perceptual span are processed in a serial word-by-word fashion or in parallel as a function of a distributed processing gradient. Nevertheless, since these models are based on research with alphabetic scripts, most of them assume word-based processing (see S.-N.Yang & McConkie, 2004, for a different view).

Evidence for word-based processing of alphabetic languages is provided by the preferred viewing location phenomenon (Rayner, 1979). The landing position distribution on a word is approximately a normal distribution slightly skewed to the left of the word center. In addition, as the distance between the prior fixation and the target word increases, the distribution shifts leftward and has greater variance (McConkie, Kerr, Reddix, & Zola, 1988). McConkie et al. interpreted this finding as evidence that a word object is selected for the next saccade whereas the actual landing position is influenced by saccadic range error and random error. The existence of the preferred viewing location implies that certain information about the next word is extracted parafoveally (presumably, because words are delimited by interword spaces, at least the location and boundaries of the next word can be obtained) and is used to determine the next fixation location.

Basic eye movement behavior during the reading of Chinese has been investigated recently. In spite of radically different orthographies between Chinese and alphabetic scripts, most eye movement patterns are similar. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.