This study examined relationships between semantic activation and eye fixation in visual word recognition of Japanese Kana and Kanji characters. In a display, a target word of either Kana or Kanji character and four images were presented, and participants identified as quickly and correctly as possible one of four locations that a corresponding target-designated image was presented. At one of the other three locations, an image which was semantically-related (Exp. 1), or an image which was graphemically-related (Exp. 2), to the target word was also presented. Additionally in the display, there were two images which did not relate semantically and graphemically to the target word. The fixation duration was obtained and the time series analyses of fixation revealed following findings. In Exp. 1, participants significantly fixated the semantically-related image longer than non-related images. In Exp. 2, there were significant differences of the time course of fixation to display images between Kana and Kanji conditions, and Kana character was assumed to be processed about 30-50 ms faster than Kanji character. It is suggested that the fixation reflects semantic activations in the visual word recognition.
