Phoneme-Based Word Activation in Spoken-Word Recognition: Evidence from Japanese School Children

摘要

This paper has investigated whether phoneme-based word activation in spoken-word recognition can be observed in Japanese by examining Japanese school children who have no knowledge of Roman alphabet. Two experiments were conducted with Japanese third graders at Japanese elementary schools with a word reconstruction task in order to test whether they could access to phonemes to reconstruct the original words from partially distorted words. The position of the distortion included both word-initial and word-medial. The results in the two experiments have clearly shown that they could reconstruct original Japanese words from the phoneme- distorted words than the mora-distorted words much faster and more accurate. The findings suggest that Japanese school children who do not have Roman alphabet may activate word candidates on the basis of phonemes, just like Japanese adults in our earlier studies and that the phoneme-based word activation may be universal in spoken-word recognition.