When talking with someone, we convey intention to each other by verbal and non-verbal behaviors. In consensus building dialogue, the participants need to understand whether they agree or disagree. They reiterate approach to partner, such as statement of opinion, and response for it. In this study, we considered that the timing of listener's response for the approach by a speaker reflects listener's internal state (agreement/disagreement), therefore analyzed the multimodal timing structures between the approach and the response by utterance and body motion. Especially, we focused on an action that the speaker turns his face toward the listener when he approaches the listener and analyzed how it influences the timing structures. As the results, we confirmed that the timing structures relate to the internal state and the relations are controlled by face-turning action.

This paper discusses acceptability of textual information and sign video as the communication method for hearing impaired people. We prepared three kinds of video for the test data; sign alone video, sign video with caption and caption alone video. Either of three video was presented to subjects. Subjects were divided into three groups; deaf native signer of JSL, deaf non-native signer of JSL and hearing people who were studying JSL. Then four Japanese written sentences, which were related to the contents of test data and included one wrong sentence, were presented. Subjects were instructed to select and answer the wrong sentence among them. While subjects were watching the test video, points of gaze were measured and recorded using an eye tracker. As the results of the experiments, deaf signers, specially native signers were good at reading sign language proportioned to that of written sentence. From the result of the analysis of eye tracking data, people who were good at reading sign language concentrated their gaze in the signer's face. mom these results, for the communication method of wide variety of hearing impared people, it is desirable to present by sign language along with by written sentences.

This paper propose a evaluation model of the website credibility. In the conventional studies it was examined the website credibility evaluation model without consider the purpose of using website. We conducted two investigations in order to reveal the structure of evaluating website credibility based on an actual web browsing. First, we conducted the questionnaire survey of clues of the website credibility evaluation to EC site and NPO site. It was extracted 3 factors, 1)basic information items,2)external evaluation information and 3)reality information. Second, we experimented to understand how these factors were evaluated by an actual web browsing behavior qualitatively. As a result, the following three tendencies were clarified.1) When as much as one necessary basic information item is lacked, the evaluation of website credibility is subtracted according to theimportance.2)"Production elements" which includes reality information, usability, design etc., can improve or lowers the evaluation of the website credibility. 3)If user feels that the website credibility is low, they tend to evaluate external evaluation information. Finally, we proposed the website credibility evaluation model with a numerical formula.

Finger Braille is one of tactual communication media of deafblind people. Deafblind people who are skilled in Finger Braille can communicate various emotions by changing durations and finger loads. In this study, we analyzed the characteristics of emotional expression by Finger Braille interpreters. And two kinds of emotional expression experiments were carried out. The one was the experiment of two-handed Finger Braille and the other one was the experiment of one-handed Finger Braille. In the experiments, two Finger Braille interpreters expressed four emotions (Neutral, Joy, Sadness and Anger) in two sentences ("Rain has fallen. " and "It is the time."). At first, we calculated the duration of code, the duration of pause and the finger load. The results of analysis of variances were:(1) the durations of code of Joy were particularly shorter than other emotions;(2) the durations of code of Sadness were particularly longer than other emotions;(3) the finger loads of Anger were particularly bigger than other emotions;(4)the durations of code in the end of sentence and clauses were particularly longer than other characters;(5)the durations of pause in the end of clauses were particularly longer than other characters.

Findings on how humans judge cultural differences in cross-cultural pictogram interpretations are reported. An open-answer web survey was conducted in the United States and Japan to collect English‐Japanese pictogram interpretations for 120pictograms used in an email system.30 pictograms with possible cultural differences were selected as stimuli for investigating human cultural difference judgment criteria. Six human subjects, three Americans and three Japanese, assessed the degree of cultural differences in 30 U.S.‐Japan pictogram interpretations using a questionnaire. Interviews were conducted to elucidate the reasons behind the judgment. The following factors were considered when judging cultural differences in cross-cultural pictogram interpretations: (i)similarity or difference of interpretations between two countries, (ii)percentage or ranking of interpretations, (iii)conformity or variance of interpretations within one country, (iv)presence of contrasting pair of proper nouns such as country names, and(v)positive or negative connotations underlying the two countries' interpretations.

Using eight hearing-impaired subjects and eight hearing subjects, this study sought the most effective point, for ease of reading, at which to break from one line into the next of captioning text presented by real-time speech-to-caption system. Results of Scheffe's paired-comparison showed that averages of preference scale of both hearing-impaired subjects and hearing subjects were high for a line-break pattern in which each line consisted of approximately 25 characters including one comma or more, and always breaking immediately after a period/comma. However, there were discrepancies in terms of degree of individual preferences. Eye movement measurements during reading captions indicated that there was a relationship between preferences for the ease of reading and how to read captioned text. Many of hearing-impaired subjects were non-follower type, i.e. many did not read captioned characters in the order of presentation. These subjects showed a high preference toward reading lines of captioning that were broken immediately after a period/comma. Moreover, fixation duration during regression of such patterns was short, even though the frequency of regression across lines was high. From these results, it is considered that hearing-impaired persons feel greater ease at reading when a captioning line breaks immediately after a period/comma. Keywords: the hearing impaired, real-time speech-to-caption system, Scheffe's paired-comparison

A psychological experiment was conducted to capture the nature of Human-Human and Human-Agent Interactions where humans and computer agents coexist. Two factors were manipulated to investigate the influences of the `schema' about the opponent and the `actual opponent's conversation' on the characteristics of communication. The first factor, expectation about the opponent, was controlled by the experimenter's instruction, manipulating with which opponent (human or computer agent participants believed to be communicating. The second factor, the actual opponent, was controlled by manipulating with which opponent (human or computer agent) participants actually communicated. The results of the experiment suggest that the degree of the refinement of the conversation controlled as the actual opponent factor affected both emotional and cognitive characteristics of communication; however the schema about the opponent only affected emotional characteristics of communication. A detailed analysis based on the variance degree of conversation suggests that the actual opponent factor becomes dominant in the large contradiction group, whereas the schema factor appears only in the small contradiction group.

InterActor is a speech-input-driven CG-embodied interaction character　that can generate communicative movements and actions for entrained interaction. InterPuppet, on the other hand, is an embodied interaction character that is driven by both speech input, similar to the InterActor, and hand motion input, like a puppet. Therefore, humans can use an InterPuppet to communicate effectively by using deliberate body movements as well as natural communicative movements and actions. In this paper, an advanced InterPuppet with a cellular-phone-type device is developed. It can be used by speech and key inputs in a mobile environment. The effectiveness of the system is demonstrated by a sensory evaluation experiment conducted in two actual remote communication scenarios.

In order to enable virtual face-to-face communication, we have developed an embodied avatar-mediated communication system by using a human avatar called Virtual Actor within the same communication space. The effectiveness of these systems has been confirmed by communication experiments. We have already proposed the concept of a Puppet Avatar in which remote talkers under the restricted conditions can operate their own avatar based on their hand motion, and developed the prototype of this system using a glove sensor. In this study, communication systems using three pointing devices‐a wireless mouse, trackball, and 3D mouse‐were developed for primitive interaction using a general PC environment. Instead of a simple trackball, a 3D trackball, which has one handle ball involving yaw rotation as well as pitch and roll rotation, was developed in order to manipulate the avatar's head motion in a more intuitive manner. A communication experiment in which a wireless mouse, trackball, and 3D mouse were compared was conducted for 15 pairs of talkers　(30 in all); a sensory evaluation and analysis of embodied avatar manipulation demonstrated the effectiveness of the proposed system.

Despite the many types of telecommunication systems that have been developed, it can still be hard to convey various types of information expressively to are mote partner. Our research focuses on the use of variations in temperature to represent information expressively. We developed a mouse device with thermal capabilities; the device becomes warmer or colder to a user's palm when the user clicks "thermal" photographic images on a computer screen. Each image has an associated thermal value. In this paper, we report the results of an evaluation of the thermal performance of the device. We also report results from a preliminary experiment that determine how the thermal expression affects a user's impression of images.

The goal of this research is to study an expression technique which creates a feeling as if the virtual tool became a part of body when a user exploited the tool. For that purpose, a novel expression is required to create a presence of a virtual tool and non-separable relationship between virtual tool and the user's body. In this research, authors pay attention to effect of shadow, because a shadow of user's body is non-separable from the user's body, and shadow enhances presence of the object. Therefore, we devised a virtual tool system which integrates two expressions of a virtual shadow of a physical tool and a three-dimensional representation of a virtual tool. Furthermore, we conducted two experiments on interacting with virtual and physical objects while using the virtual tool: a questionnaire method and a unique method based on cross modal interference. Two experiments demonstrated that our virtual tool has an effect on creating a feeling as if the virtual tool became a part of body.

It's said that embodied coherence plays a significant role in human communication. But, in what situations an embodied coherence is generated is as yet not well known. So, in this research, we empirically investigate in what situations embodied coherence generates on condition that participants always continue to have cooperative relationship. Specifically, we developed experimental system, in which players operate their own cursor linked to other cursor with the rhythm controller and prevent from a collision with an obstacle(ball)in the virtual space. As a result, we found that participants tend to generate the embodied coherence before ball approaches their cursors. Furthermore, we also found that the situation that it is difficult to anticipate partner's cursor movement leads to a higher frequency of generation of the coherence. We show that the embodied coherence is generated for share of the anticipation when participants anticipate the future from the past situation.

When a small object in a hand-held device moves periodically and prismatically with asymmetric acceleration (strong in one direction and weak in the other), one typically experiences the kinesthetic illusion of being pushed or pulled continuously by the held device. This effect was investigated because of its potential application to a hand-held, non-grounded, haptic device that can convey a sense of a continuous translational force in one direction, which is a yet missing the in haptic research. Here, a one-degree of freedom haptic device based on a double-layer crank-slider mechanism was constructed based on the results of our previous research. We evaluated the effectiveness of the perception of directed force sensation by asymmetric oscillation compared with prior prototype, and the effectiveness of the gross weight of the device. Our results show that the ratio of the gross weight of the device and the weight of reciprocating mass should be at least 16%for effective force perception.

The force-feedback in master-slave system is effective to get comfortable maneuverability, and has often been introduced in various applications. In conventional systems, when the slave robot is in free space, no reflective force from the slave robot's environment is returned to the operator, because there is no contact between the slave robot and the environment. Contrary to the conventional method, the authors proposed a method to present the kinesthetic sense as a virtual force even though there is no con-tact between the slave robot and the environment. The virtual force reflects the slave robot's situation such as position and velocity, and the reflection force helps the operator to maneuver the slave robot in the di冊cult environment with the favorite and/or dangerous regions. Because the intervenient dynamic model governs the maneuverability and the stability of the system, a designing method considering the intervenient dynamic model has been required. However, no practical designing method has not yet been proposed for the virtual force-feedback system. This paper proposed a method for designing a dynamic model of the master-slave system with virtual force-feedback. The effectiveness of the proposed method was confirmed through some practical experiments.

In this research, the concept of the hybrid information space that integrates the information seamlessly between the virtual world and the outdoor field is proposed, and the prototype system was developed. In this system, a user can access information stored in the database using the cellular phone, PC and the immersive projection dis-play as information devices. The virtual world is constructed by placing the photograph images that were captured in the outdoor field according to the positional information measured by GPS function of the cellular phone. This system can be applied to the prior and posterior learning for the study tours, by experiencing the virtual world before and after going to the outdoor field as well as accessing information using the cellular phone. This system can also be used for the communication between the outdoor field and the virtual world while sharing the spatial information.

As portable electronics and various types of devices are constantly being downsized, the style of operation is being diversified. The control strategy in situations where the equipment controls are not situated in front of a human body or face is very different from a face-to-face situation. It is therefore important to clarify the mechanism of direction judgment when the equipment controls are not situated directly in front of the human body or face. The purpose of this study was to investigate the effect of the orientation of an operator's hand movement based on direction judgment in a cognitive manner. In our experimental conditions, right-handed participants were asked to judge directions by the voice instructions provided by the computer through certain sound sand to push the appropriate button that corresponded to the direction in a cognitive manner. In a series of three experiments, we verify the effects of the orientation and positions of an operator's hand, and the differences between the dominant and non-dominant hands on control performance. The results of these experiments indicate that the ease in mental imaging of the control direction is particularly influenced by hand orientation. In addition, the results also show that for control strategy, there are differences between dominant and non-dominant hands.

We have developed a system named "Wide-view electronic working space "which realizes a gesture action combined with a wide-view arch display. After the use of the system, there were many demands in which more than two people can operate and cooperate in the system. In this study, we designed and implemented aside-by-side system for multi-user cooperative works with a wide-view working space. In addition, we realized video editing application by sharing and cooperating works, which was able to evaluate the utility of the side-by-side system.

A new pen device called data-embedding pen is proposed, which can embed digital data into a handwriting trajectory as a small ink dot sequence printed along the trajectory. The produced handwriting patterns convey not only usual human-readable information by the handwriting trajectory but also machine-readable information by the ink dots and therefore have a potential for a universal man-machine interface. In this paper, after discussing hardware and application of the data-embedding pen, component technologies such as extraction and order estimation of embedded ink dots from hand-writing images and error correction, are detailed with experimental results.

The present paper reconsiders the active four-wheel steering controller and the steering reactive torque presentation method, which were proposed in the previous studies, from the viewpoint of an ecological interface design. It shows that drivers can intuitively understand how the control law works and the haptic interface enables drivers to perceive the vehicle state. Moreover, an ecological multimodal interface, which presents visual information about longitudinal and lateral accelerations based on the concept of friction circle in addition to the haptic information, is proposed to inform the road surface condition and the working state of the vehicle control system intuitively and comprehensibly. Driving simulator experiments verify the effectiveness of the proposed multimodal interface.

Social Networking Services(SNS) are now popular, and there are several activities in which they use SNS to accumulate users' idea and to share knowledge. We have been working on a method to store and share "Learning Process" on SNS, in order to bring about more effective learning. In this method, a student usually posts articles or observations in their individual space, and other students are free to post comments. In this research, we propose an interface to visualize these articles accumulated in users' individual spaces. This interface features a zoom-able user interface and an overview of all users' articles. We can then evaluate the usability of the interface to view data accumulated in SNS through practical learning activities that we held so far. Experimental results have indicated that a proposed interface is more easily accessible and effective than existing SNS interfaces. Also, students have been using it with a high degree of satisfaction.