Effective visual field size necessary for proofreading during Japanese text editing

Abstract

Effective visual field size and reading time necessary for visual search was measured during proofreading of Japanese text. Proofreading performance in detecting erroneously printed characters in both hirakana and kanji was measured using a variable moving window, through which the subject could read the areas of the text. The moving window based on the current eye position, was generated in real time on a computer-controlled CRT screen, determined by an eye movement recording system (600 points/sec). Movement of the window in which text was presented was dynamically controlled. Subjects were asked to detect transposed characters (Mozer, 1983) both in hirakana and kanji (words were composed of 2 to 4 characters). Detection of erroneously printed words was found to be relatively easier in kanji than hirakana. The effective visual field size tended to change depending on the type of script. Furthermore, as the window size increased, the proofreading rate increased significantly. Individual differences between effective visual field size and reading span score were also investigated. The subjects were divided into high and low score groups using reading span tests, which can measure working memory efficiency during Japanese text reading (Osaka & Osaka, 1992). The results suggest that effective field size tended to increase for subjects with higher scores and in general, reading performance was better for subjects with higher scores. The results are discussed in terms of working memory efficiency and field size.