skip to primary navigation skip to content

CBSU bibliography search

To request a reprint of a CBSU publication, please click here to send us an email (reprints may not be available for all publications)

Phonologically mediated access to meaning for Kanji: Is a Rows still a Rose in Japanese KANJI?
Wydell, T.N., Patterson, K.E. & Humphreys, G.W.
Journal of Experimental Psychology: Learning, Memory, and Cognition, 19, 491-514.
Year of publication:
CBU number:
It is generally assumed that access to phonology for logographic Japanese KANJI characters is mediated by access to their meaning, and that this semantically mediated route is the only way to retrieve the phonology of KANJI words. This proposal was examined utilising a semantic judgment task with homophones, as developed by Van Orden (1987). Subjects were asked to judge whether or not a target word was an exemplar of the category name they had just seen (e.g., in English, given the category name FLOWER, is "ROWS" a correct exemplar?). If access to KANJI phonology is always mediated by access to meaning, then homophony should have no effect on semantic judgments. However, the results showed that there was (1) a significant homophony effect - RTs were longer and more errors were made with homophone foils than with control foils, (2) a significant effect of visual similarity both with homophone and control foils - RTs were longer, and more errors were made, when the incorrect target words were visually similar to correct exemplars of the category names, (3) a significant interaction between homophony and visual similarity - there was no effect of homophony with visually dissimilar homophone foils, and homophony increased the effects of visual similarity, (4) a significant interaction between word frequency and visual similarity - most errors were made on visually similar foils corresponding to low frequency correct exemplars, and (5) a significant visual similarity effect on errors even under visual masking conditions. The data suggest that the reading of KANJI is characterised by parallel access to semantics from orthographic and phonological representations.