漢字の形態処理, 音韻処理, および意味処理の関連性について 形態マッチング課題を用いて

井上 道雄

doi:10.4992/jjpsy.51.136

漢字の形態処理, 音韻処理, および意味処理の関連性について

形態マッチング課題を用いて

井上道雄

著者情報

ジャーナルフリー

1980 年 51 巻 3 号 p. 136-144

DOI https://doi.org/10.4992/jjpsy.51.136

詳細

発行日: 1980/09/20 受付日: 1979/07/09 J-STAGE公開日: 2010/07/16 受理日: - 早期公開日: - 改訂日: -
訂正情報
訂正日: 2010/07/16 訂正理由: - 訂正箇所: 論文抄録訂正内容: Wrong : The purpose of the present study was to investigate the relations among the graphemic, phonemic, and semantic processings of Chinese characters (KANJI). Specifically, it was to define what kinds of processings are dominant, or how do those processings function interactively in word recognition processes.
In this experiment, the graphemical matching task was performed to 20 university students. The task was to decide whether or not a stimulus pair was graphemically identical, and the subjects made either “Same” or “Different” judgement by releasing a right or left response key as fast and accurately as possible. Stimulus pairs were constructed from two-letter Chinese characters (JUKUGO), or a compound word. Each stimulus of a pair was serially presented on the screen for 100ms each.
Three factors, i.e., interstimulus interval (ISI), word frequency, and type (only for Different-judgement) were manipulated in the experiment as follows. As to the ISI and frequency factors, there were two conditions each -200ms and 2000ms, and High and Low. As to the type factor, which signified the relations of a stimulus pair in the three processings, there were five conditions (see Table 1) -(a) Homophone-Synonym pair (Ho-Sy), (b) Homophone pair (Ho), (c) Synonym pair (Sy), (d) Identical first letter pair (IF), and (e) Unrelated pair (UN). The UN condition served as a control condition for the Ho-Sy, Ho and Sy conditions, and the IF condition for the Ho-Sy condition in the graphemic relation.
The main hypotheses proposed in this study were as follows: If a stimulus pair is matched on the base of each processing, the relations among processings will be manifested as interferences in Different-judgement. This interference effect is defined as the differences of matching speed between the UN and other conditions. If those three processings function interactively, the greatest interference effect will occur in the Ho-Sy condition. However, if they function independently, the interference effect will also occur in both the Ho and Sy conditions.
The results and conclusions are as follows. First, the results on the type factor in Different-judgement showed that the interference was in the Ho-Sy than IF condition, but it did not occur in the Ho and Sy conditions (Fig. 2). These results suggest that the three processings function in parallel and interactively in the word recognition processes of Chinese characters even in the graphemical matching task. However, it also suggests that the phonemic or semantic processing does not influence independently on the graphemical matching.
Secondly, results on the ISI factor showed that the matching speed was faster in the ISI 200 than ISI 2000 conditions in both Same and Different-judgements, and those patterns of reaction time were constant in all conditions of the type factor (Figs. 1 and 2). These results suggest that three processings are carried out in parallel within 300ms, and the shift of dominance among those processings or the change of codes is not occurred at least within 2100ms.
Finally, the frequency factor hardly influenced the matching speed in Same-judgement, but it did for the Ho-Sy and IF conditions in Different-judgement. It thus appears that only if there are relations in both the graphemic and phonemic processings, the semantic processing will influence the graphemical matching.
Right : The purpose of this study was to investigate the relations among the graphemic, phonemic, and semantic processings of Chinese characters (KANJI) in word recognition processes. The graphemical matching task, which was for subjects to decide whether paired stimuli presented serially are graphemically identical, was performed to 20 university students. Three factors, which might influence the reaction time in the matching task, were munipulated. They were interstimulus interval (200 ms and 2000 ms), word frequency (High and Low), and type (five conditions as to the relations of a stimulus pair in the three processings). The results were interpreted as suggesting that these three processings functioned in parallel and interactively, that they were carried out within 300 ms, and that the change of codes was not occurred within 2100 ms.

抄録

The purpose of this study was to investigate the relations among the graphemic, phonemic, and semantic processings of Chinese characters (KANJI) in word recognition processes. The graphemical matching task, which was for subjects to decide whether paired stimuli presented serially are graphemically identical, was performed to 20 university students. Three factors, which might influence the reaction time in the matching task, were munipulated. They were interstimulus interval (200 ms and 2000 ms), word frequency (High and Low), and type (five conditions as to the relations of a stimulus pair in the three processings). The results were interpreted as suggesting that these three processings functioned in parallel and interactively, that they were carried out within 300 ms, and that the change of codes was not occurred within 2100 ms.

責任著者(Corresponding author)

訂正情報

J-STAGEへの登録はこちら（無料）