Previous research studies on supporting EFL (English as a Foreign Language) reading showed that the Kit Build concept mapping (KB-mapping) method provided better efficiency than the Scratch Build concept mapping (SB-mapping) method when evaluated by a delayed comprehension test two weeks after the reading activity, although there was no difference in test results immediately following the reading activity. In the present study, we set out to investigate the following research question “Why does the KB-mapping method have better efficiency than the SB-mapping method in recalling the comprehended information two weeks later although there is no difference just after the reading?” We had observed that there was a difference in map-building behaviors between KB-mapping and SB-mapping methods, so we made the assumption that “KB-mapping method does not promote building the map sentence-by-sentence, although the SB-mapping method does”. To confirm this assumption, we used an experiment to compare the learners' map-building progress by the KB-mapping and SB-mapping methods. In the experiment, we monitored the performance of learners in terms of the size of the maps and the progress they made for building maps. We added new functions to our system to record the building process of every learner. We observed that SB-mapping learners tended to use the sentence-by-sentence map-building style, but KB-mapping learners did not. Although there have been no previous research studies about map-building style in supporting reading comprehension with concept mapping, many researchers have indicated that sentence-by-sentence reading is not a good way to comprehend content deeply and keep it in memory for a long time, although sentence-by-sentence reading is useful to understand content rapidly and to keep it for a short time. In map-building, we suspected the same disadvantages and advantages would be true, and this would explain the results, KB-mapping method helps learners to avoid the sentence-by-sentence map-building style, but the SB-mapping method cannot help learners escape from this style of map-building. Based on these results, we will design adaptive support for reading comprehension in our future work.

View full abstract

In this paper, we propose an embodied conversational agent based on the willingness to communicate (WTC) model in L2 to help increase WTC in the context of English as a foreign language (EFL) by providing users with various daily conversation contexts. To simulate realistic and efficient conversations, we adopted a semantic approach in the response generation and created a system with flexible and adaptable domain knowledge, user's intent detection, and mixed-initiative conversation strategy. Our evaluation of the proposed system demonstrated its potential to simulate efficiently natural conversations in a specific context as well as the feasibility of improving learners' WTC using a computer-based environment.

View full abstract

We discuss how a system to support the facilitation efforts of inexperienced case leaders can be brought close to reality. We begin by considering methods to support facilitation efforts directed at novices. For this purpose, we conduct two tests. Using the test results, we propose that as one capability of a support system, the system should include some mechanism for recommending participants found by pretesting to already have a high level of knowledge, and that such recommended participants be given particular encouragement by the facilitator to externalize their discoveries. Also, the system should include some mechanism for recommending how to split up group members based on pretest results. By this mechanism, we can provide each group with a good balance of participant skill levels, including participants with a high skill level, and such an arrangement should encourage the groups in learning occurrence by letting them discover a new viewpoint.

View full abstract

Although summarizing English texts is an effective method for improving English speaking skills, it is difficult for learners to use this approach through self-study without teachers' support. In this study, we developed the Mobile Application for Listening and Scaffolded autonomous summary speaking (MALS) and evaluated its effectiveness for increasing oral fluency. MALS offers not only a summary speaking task but also scaffolding practice that decrease the cognitive load of summarizing by facilitating learners in confirming the summary points and the recommended words. The results showed that (1) the scaffolding enables learners to do a summarizing task in their self-study, and (2) MALS is effective for not only increasing the number of non-repeated words and decreasing silent time but also decreasing frequency of repair. Hence, MALS compares favorably with reading aloud activities that are judged effective for only increasing the number of non-repeated words and decreasing silent time.

View full abstract

The Web allows learners to investigate any question with Web resources to construct wider and deeper knowledge. In such an investigative learning process, it is important for them to elaborate their question, which involves decomposing it into the sub-questions to be further investigated. Such question decomposition corresponds to creating a learning scenario that implies the questions to be investigated and their sequence. However, it tends to be insufficient. This paper addresses the issue of how to promote the question decomposition. Our approach is to model the Web-based investigative learning process. In this model, the learning scenario is represented as a tree of questions to be investigated. We have also developed an interactive learning scenario builder (iLSB), which provides some scaffolds for building the question tree as modeled. This paper also reports results of our case study suggesting that iLSB has a potential to make an elaborate investigative learning process.

View full abstract