Previous researches suggest that STEAM education contains both STEM-based and art-based ones. The purpose of this paper is to describe a design of collaborative creative activity in music and an evaluation method as a means of considering art-based STEAM education which improves students’ self-esteem. We designed a three-hours musical workshop for 20 children who were five graders in an elementary school to facilitate the creation of harmonies to the given melodies. The children used ICT to externalize their ideas in collaborative learning, and they were evaluated pre-, intermediate- and post-test. Each test contains a questionnaire, an exam of creating harmonies, and questions of self-esteem. The results of the questionnaire indicated that all the children enjoyed the workshop and some of them became externalization than appreciation. Furthermore, the children who made not qualified harmonies improved their self-esteem in post-test than pre-test, though the self-esteem of the children who made moderate quality of harmonies decreased. Detailed protocol analysis suggested that the pair of children who improved their self-esteem allocated their time to consider each other’s ideas and not hurried to complete making their harmony. The importance of evaluating the children’s internal change in this workshop as an art-based STEAM education was discussed.

抄録全体を表示

With the development of internet technology, people can share knowledge through Q&A websites anytime and anywhere. However, two types of barriers between the questioners and respondents make knowledge sharing difficult: 1) lack of trust for other users, 2) an arrogant or negative attitude exhibited by the questioners. We conducted an experiment to evaluate the effect of the virtual agent for lowing these barriers with two characteristics: 1) similar profile to users, good appearance and affable personality for users, 2) positive attitude and reaction. The results indicate that the virtual agent can improve the users’ motivation to answer questions.

抄録全体を表示

Studies of human-human interaction have not focused on consideration in approach. This is obviously very important if people are going to interact with a variety of people. In this study we closely observe the way people behave and interact based on consideration of others. More specifically, the study describes a series of experiments in which human participants approach a mannequin who is apparently concentrating to some degree on reading a book, then conduct the task of asking the mannequin for directions. As a result, when participants see that the other party is absorbed in something, they feel hesitant to approach front-on and speak to the person directly. Rather, they tend to circle around to the side to initiate conversation with the other party. This behavior does not encourage the other person to speak, but rather brings one in closer to the other party and creates a congenial conversational situation to initiate conversation. Our experiment demonstrates a tendency for people to employ subtle psychological strategies, even if nonverbal, to work one's way into the other person's state.

抄録全体を表示

In an evacuation behavior analysis, records of behavior such as straying are dependent on participants' memory. Therefore, there is a loss of some of behavior records. In this research, we propose pair-testing method for supplementing records. A participant pair with an observer and records the behavior of participant. As a result of evacuation experiment, the behavior of participants was supplemented by the observer by 47.5%. It is expected that evacuation behavior will be verified from the relationship between these behavior data and context-dependent data.

抄録全体を表示

How can we communicate with people who don’t share a common language? To explore this question, this paper describes a computational model that simulates a experimental study. In the experiment, we employed a coordination task in which participants used a novel communication media that restricted usual communication methods. The model was developed to build behaivor charactersits observed in the experiment from repeated interactions of ACT-R agents. We focus on the cognitive mechanisms implemented in ACT-R architecture such as instance-based learning, imitative learning, and implicit memory process provide the scaffold for building a common communication system. From a simulation study manipulating these mechanisms, we discussed factors establishing a common communication system.

抄録全体を表示

This research is the first attempt to analyze how the gaze behavior during turn-keeping/changing changes depending on empathy skill, which is measured by using Davis' Interpersonal Reactivity Index (IRI). The representative results of the analysis of the gaze behavior and empathy skill showed that current speakers and next speakers with a high empathy skill level usually do not look at other people during turn-keeping or turn-changing. In contrast, speakers with a low empathy skill level often continue to make eye contact with listener during turn-keeping and turn-changing. Moreover, listeners with a low empathy skill level often look at the speaker unilaterally during turn-keeping and do not look at others during turn-changing. These are new findings. On the basis of these findings, we constructed four estimation models of the four-scale value of empathy skill levels by using gaze behaviors. The evaluation results showed that the average absolute error of estimation is only 0.22 for the gaze transition pattern (GTP) model.

抄録全体を表示

In this study, we focus on solving the problem of gaze presentation in telepresence robots with a rotatable display that causes lack of presence or low participation of remote participants. As a role that gaze should play, we focus on "mutual gaze" and "joint attention" and present a system to fulfill these roles. We implemented a new telepresence robot system which used the humanoid head as a physical pointer, and conducted two experiments. Results of these experiments reveal that the proposed system with humanoid head can increase the accuracy of gaze direction estimation, and also improve sense of “being looked at”, “co-presence”, and “face to face”. Furthermore, the mutual gaze is accomplished more often with humanoid head. The results also reveal that not only proposed system but also the system with display rotation are effective to guide participant’s attention and improve impression of joint attention.

抄録全体を表示

Touchscreens are common in personal devices as well as in public devices such as ATMs and vending machines, because they are customizable and require less mental and physical load. However, touchscreens are subject to human error and visual overload, because a touchscreen interface doesn’t have haptic clues. Furthermore, the elderly may tend to be more affected by these issues. We conducted the experiment by measuring finger movement and finger contact force while using a touchscreen, to ascertain suitable feedback from haptic clue for both the elderly(62-75 years old) and younger people(20-30years old) . The findings were as follows: The amount of finger movement and finger contact force of elderly participants were significantly more than that of the younger participants (p < 0.01). Increased finger movement was due to tremors and decreased dexterity caused by aging. Increased finger contact force was due to two reasons: first, a response to less haptic feedback than what elderly participants are accustomed to experience, such as in the operation of a mechanical button; second, an attempt to prevent an error in movement due to their tremor. In addition to the physiological measurements, interviews of the participants revealed that the reason for their differences in using ICT devices is not a lack of experience, but physiological issues, such as tremors, decreased sensation in aging. The study further confirmed that touchscreen designs should include larger buttons and text display and a haptic experience that is attuned to strong finger contact force to promote greater usability for elderly users.

抄録全体を表示

The purpose of this study is to evaluate an acceptable dwell time range for object selection using mid-air hand gesture interaction system. In such a system, a user selected a target object by moving his/her palm and dwelling on it. We focused on the situation that there were densely arranged objects and the user selected a target object among these objects. An experiment was conducted to evaluate the effect of dwell time on the object selection task. The results indicated that the lower limit of the dwell time range is about 0.2 s to 0.5 s, depending on the distance between objects, although the upper limit was not clear. The contribution of this work was expected to be the basis of the dwell time for selecting operation in mid-air hand gesture interaction system.

抄録全体を表示

Thanks to the rapid progress of 3D printers, we can easily print 3D objects that we made ourselves. However, it is still difficult to freely engrave marks on the 3D virtual object's surface in a similar way to the real carvings. In this study, we propose a virtual carving system in a Mixed Reality (MR) space which the users can engrave marks on 3D virtual object’s surface. In our system, by touching and tracing CarvingToolDevice on the real object, users can carve on the 3D virtual object's surface which is superimposed on the real object. This paper describes the design and implementation of the proposed system and describes the findings through the user's experience.

抄録全体を表示

Metacognition is essential for learning. For the learners, writing activity with the expert is a great opportunity to acquire the metacognitive knowledge by inferring the expert’s metacognitive processes for critical reading. In this study, we define “metacognitive inference activities” as inferring experts’ metacognitive thinking processes. In order to support learners’ metacognitive inference activities, we analyze the potential of eye-movements information as a novel learning material to promote learners’ critical reading of their own academic paper. Based on learners’ and experts’ eye-movement information, the proposed system provides two different types of visualization: “comparative heat map view” which shows the differences between total gazing-time of respective sentence-objects of the learner and those of the expert, and “eyemovements visualization view” which represents the untouched expert’s eye-movements processes during his/her critical reading. Experimental results showed that the visualized information of the comparative heat map and the experts’ eyemovements processes have the potential to promote learners’ metacognitive inference activities towards being aware of metacognitive knowledge without instructions.

抄録全体を表示