In order to provide an effective feedback, we have performed the classification based on the articulatory features in terms of both the place of articulation and the manner of articulation. Figure 4.4 shows an example of consonant feedback for learner B while he mispronounced /d/ for /t/ in [kyooto]. In addition, he did the same mispronunciation for the other data in the T-set: [wadashi] for the correct pronunciation [watashi]. That is to say, he showed the pronunciation problem in producing voiced dental plosive (/d/) for the target sound unvoiced dental plosive (/t/). Actually, this kind of pronunciation error could be easily corrected if the learners became aware their mistake through feedback instruction.
However, for the pronunciation errors which are intrinsically hard for them to pronounce due to the linguistic gap between their L1 and target L2, we need to adopt a different strategy. Figure 4.5 shows the classified results by the place of articulation and the manner of articulation for the phoneme /sh/ for learner B, i.e., shiNyou, moushikomu, shurui, shourai, aNshiN, shiNkou, shochi, shuukyou (the P-set). The results show that he tends to pronounce unvoiced dental not post-alveolar for /sh/ for the place of articulation. On the other hand, his pronunciation for the manner of articulation was appropriate.
