2015 Fiscal Year Research-status Report
第二言語使用時における無声・有声休止と文構造の関連性
| Project/Area Number |
15K02765
|
|---|
| Research Institution | Waseda University |
|---|
Principal Investigator |
Rose Ralph 早稲田大学, 理工学術院, 准教授 (30404916)
|
|---|
| Project Period (FY) |
2015-04-01 – 2018-03-31
|
| Keywords | 第２言語習得 / コーパス言語学 / 無声休止 / 有声休止 / silent pause / filled pause |
|---|
| Outline of Annual Research Achievements |
The focal point of the first year of the project was a corpus study of the relationship between silent and filled pauses and syntactic structure as evidence by clause length and clause depth. Much of the year was spent on adapting and annotating the corpora to be used in the study. The corpora included a speech corpus of native English speakers as well as a crosslinguistic speech corpus of natives speakers of Japanese speaking in both Japanese and their second language, English. In both corpora, annotation of clause structure was necessary. For both corpora, a simple measure of clause structure as clause length was obtained through the annotation process. However, a more sophisticated measure of clause structure as clause depth could not be completed fully for both corpora and will be continued in year two of the project.
The results of the corpus annotation show some interesting differences between the two corpora and hence between the two groups of speakers. In short, as second language learners become more proficient in the target language, their pausing (both silent pauses and filled pauses; uh/um in English, e-to/ano in Japanese) rates and durations as well as utterance length more closely resemble those of natives. However, the interaction of these features -- i.e., their pausing strategies -- may still show a very different production profile.
|
| Current Status of Research Progress |
Current Status of Research Progress
3: Progress in research has been slightly delayed.
Reason
Part of the plan for completing the corpus annotation of clause structure involves completing some of the work using crowd-sourcing methods. However, the preparation of the platform for doing this work is quite time-consuming. By the time it was completed, it was too late in the project year to execute the crowd-sourcing plan effectively with proper and careful monitoring of the incoming results. Therefore the decision was made to delay it until the second year of the project.
|
| Strategy for Future Research Activity |
The project will continue by first completing the corpus annotation of clause structure using a crowd-sourcing approach. This will permit the completion of the corpus study stage of the investigation and set the stage for the perception experimental study that is planned for the next stage. The experiment will involve a large number of participants listening to English sentences containing certain configurations of silent and filled pauses in variance with different clausal structure to represent complexity. Participants will be required to listen and respond to the sentences in a specific way that will allow a measurement of their response to the manipulations. These experiments will be performed in a sound-proof room (to prevent distraction) using a specialized computer-based presentation and response detection system. Depending on feasibility, an alternative (or possibly concurrent) approach may use crowd-sourcing methods to carry out the experiment on-line. Because of the cross-linguistic nature of the study, the former approach may be used to test native Japanese speakers locally, while the latter approach may be used to test native English speakers remotely. If the latter approach is not feasible, then travel to the US will be arranged to carry out the experiment with native English speakers abroad. The results of these experiments will be reported at domestic and international conferences, as scheduling permits. Also, in order to cultivate and develop research around the core topic of pauses and syntactic complexity, a workshop will be sponsored and held in Tokyo.
|
| Causes of Carryover |
As explained above, the corpus annotation procedure using crowd-sourcing methods was delayed at the end of the first year of the project. The second year of the project will allocate the carry-over funds to this work. As for the funds to be budgeted for the second project year, the costs of managing the experiment and providing remuneration to the participants as well as the experiment administrators will be required. Finally because the current value of the yen, international travel expenses are particularly high and will be required both for the primary investigator's and the related researchers travel.
|
| Expenditure Plan for Carryover Budget |
The budget for the project's second year will be spent in several ways. First, funds will be required to complete the annotation of the corpus by crowd-sourcing methods. Second, funds will also be required to carry out the experimental plan. This includes funds for paying the actual experimenters (graduate-level research assistants) as well as remuneration for experimental participants. Finally, a significant portion of the budget will be required for travel expenses to domestic and international conferences, as well as travel costs to bring a speaker for the workshop.
|
