Lack of semantic parafoveal preview benefit in reading revisited

Keith Rayner, Eilzabeth R. Schotter, Denis Drieghe

Full Text: PDF   Paper Package: RaynerSchotterDrieghe2014_1.0 tar.gz PID: 11022/0000-0000-1F90-B


In contrast to earlier research, evidence for semantic preview benefit in reading has been reported by Hohenstein and Kliegl (2013) in an alphabetic writing system; they also implied that prior demonstrations of a lack of semantic preview benefit needed to be re-examined. In the present article we report a rather direct replication of an experiment reported by Rayner, Balota, and Pollatsek (1986). Using the gaze-contingent boundary paradigm, subjects read sentences that contained a target word (razor), but different preview words were initially presented in the sentence. The preview was either identical to the target word (i.e., razor), semantically related to the target word (i.e., blade), semantically unrelated to the target word (i.e., sweet), or a visually similar non-word (i.e., razar). When the reader’s eyes crossed an invisible boundary location just to the left of the target word location, the preview changed to the target word. Like Rayner et al. (1986), we found that fixations on the target word were significantly shorter in the identical condition than in the unrelated condition, which did not differ from the semantically related condition; when an orthographically similar preview had been initially present in the sentence fixations were shorter than when a semantically unrelated preview had been present. Thus, the present experiment replicates the earlier data reported by Rayner et al. (1986) indicating evidence for orthographic preview benefit, but a lack of semantic preview benefit in reading English.

Psychonomic Bulletin & Review