Abstract:

We have been doing research on an effective training method for Japanese speakers to be able to discriminate between the Mandarin Chinese nasal codas “-n” and “-ng”, which is one of the most difficult phonemic contrast for Japanese speakers. The previous studies mainly focused on the characteristics observed in the process of perceiving nasal codas, and tried to investigate the error patterns caused by Japanese speakers. Few research studies have focused on a solution specifically for Japanese learners of Mandarin Chinese to be able to discriminate between “-n” and “-ng”. Based on this background, we are working on developing ICT (Information and Communication Technology) educational software for perceptional training of Mandarin Chinese nasal codas. In this presentation, we will first introduce our perceptional training system, and then, describe the results of the preliminary training tests with and without our training system.

The training software we have developed runs on the Windows operating system as a GUI (Graphical User Interface) program. Users are instructed to listen to a Chinese syllable and to identify the word final consonant in each item. There are two modes: a test mode and a practice mode. Both modes allow users only to click buttons. In test mode, users are not allowed to listen again. After the answer selection, users proceed to the next test item. At the end of the test mode, the number of correct answers is displayed. In practice mode, unlike the test mode, users are allowed to replay the same sound as many times as they need, and are also given feedback after each selection to understand whether the sound they heard was “-n” or “-ng”. Users can also proceed to the next or previous training item. It is possible to practice the same training item again.

To verify the usability of the training software, we asked Japanese speakers to train their perceiving ability to distinguish between “-n” and “-ng”. We divided the participants into two groups; an experimental group and a control group. Learners assigned to an experimental group were required to train themselves with our software. The experimental group obtained a higher score in the identification test than the one before training, while the score of the control group remained the same. Moreover, by observing the participants’ behaviors while training with the system, and their opinions, we found that there were no problems in system usability. These facts indicate that so far our training software has been effective. We are planning to verify the effect of the training software by conducting a large-scale user test.

Keywords:

Japanese speaker, nasal codas, Mandarin Chinese, training software.