h_journal
Online edition:ISSN 2434-3404

t_rules

Why do Japanese aphasics have difficulties in writing both Kana nonwords and words consisting of even two letters? *

  The dual-route model assumes a nonlexical route passing through the auditory analysis system, the phoneme-grapheme conversion system and the orthographic buffer. The Japanese writing system consists of kana(phonograms)and kanji(ideograms). When a Japanese is asked to write a kana nonword, it is processed by the nonlexical route according to the dual model. When a Japanese aphasic patient who is able to write all elements of kana is asked to write a kana nonword, theoretically this aphasic should be able to write kana nonwords correctly because the phoneme-grapheme conversion is automatically processed. Clinically, however, we sometimes experience Japanese aphasics who are able to write kana elements but who cannot write a nonword correctly, even one consisting of two letters. The purpose of this study was to test whether the dual route model can predict the accuracy of serial letter writing in Japanese aphasics. We asked 16 Japanese patients with aphasia to write both words and nonwords consisting of a single or two-to-five kana letters. The results showed that (1) the accuracy of two-letter writing was significantly lower than that calculated from the accuracy of one-letter writing based on the dual route model, (2) the accuracy of writing of the second, third, fourth, and fifth letters of both kana nonwords and words was lower than that of the first letter, and (3) patients could write serial letters more accurately when asked to write one letter per one sheet of paper than when asked to write all letters on one sheet of paper. Our results suggest that Japanese aphasics have damage of the orthographic buffer where graphemes converted from phonemes are temporarily stored. Since the buffer is thought to be a working memory system, there should be a length effect. Our results, however, suggest that the graphemic representations do not consist simply of linearly ordered sets of graphemes. The orthographic buffer may be organized so thst the letter of a word or nonword is not processed independently, and therefore the adjoining letters interfere when writing it. (Accepted on September 20,2007)

Author
Miyazaki Y
Volume
33
Issue
4
Pages
297-305
DOI
10.11482/2007/KMJ33(4)297-305,2007.pdf

b_download