Making smart speakers proactive might benefit users by delivering unconscious but useful information. However, depending on the timing and manner of the delivery, proactive deliveries pose a risk in disturbing user’s ongoing activity. Thus, we hypothesized that the times of activity-transition are appropriate for information delivery. In this study, we prototyped a system that estimates activity-transitions based the user’s body motion using a depth camera, and actively provide Internet news through a voice of virtual character. We then conducted a set of experiments in everyday-living scenario at the homes of five university students living alone. The result demonstrated that the times after transfer are generally appropriate. In contrast, the change of the user’s face direction, i.e. supposed gaze-target transition, occurred more frequently but included more inappropriate cases. Elimination of the error-detections such as body posture change while looking at smartphone is needed to detect activity-transitions more appropriately.
In machine translation-mediated communication, when each party has different cultural and language backgrounds, a particular word could be related to different meanings and different images. This could cause failures to establish mutual understanding. To solve this problem, we have proposed an image feature-based method to automatically determine words that could cause misunderstanding. This method calculates and determines the optimal threshold by comparing the result from the automated method to human judgment. We applied this method to 500 concepts and compared the judgments using 400 concepts for threshold optimization and 100 concepts for testing our proposed method. We found that 0.55 was the optimal threshold with 76 percent accuracy. Moreover, we conducted the chi-square test to determine whether the accuracy is significantly different among hypernyms of the concepts and the result statistically did not recognize any significant difference.
In this study, first, we analyze the relationship between personality traits and the expressivity of hand gestures in dyad interaction. Second, based on the analysis results, we propose a method for agents’ gesture generation that can express their personality traits. Our user study reveals that expected personality traits can be perceived from the agent’s animation generated by our proposed method. Especially for extroversion and emotional instability, agent gestures generated based on our method successfully gave the expected impression to the human subjects.
In this paper, we investigate an interactive read-aloud method for a news caster agent that gives the impression of familiarity, trust, consideration and kindness for users without the need for private conversations. In recent years, CG avatars and virtual agents get more popular, and interactive information services using anthropomorphic media have been developed. Virtual agents have the advantage of increasing intimacy with users through their anthropomorphic appearance and voice communication. On the other hand, agents providing public news information cannot take advantage of the characteristics of virtual agents because the provider avoids excessive personalities of the agent and private communication between users and the agent in order to give users an impartial and neutral impression. In this paper, we propose a method to reading-aloud-news by virtual agents, which makes users feel familiarity, trust, consideration, kindness, and close to them without private communication. We designed the interactive reading-aloud-news method for the personal newscaster agent; the agent repeats the same sentence with easy words or slow and loud voice (or with both manners) corresponding to the user’s difficulties in understanding or/and hearing of the sentence. We evaluated our proposed interactive reading-aloud-news method from the viewpoints of the following three; 1) how the user interprets the reading manners into the agent’s intention, 2) how the reading manners affect the impression of agent and 3) how the reading manners affect the im pression of news. The result of evaluation shows that the reading method makes the user feel as though the agent tried to understand the user’s state and elevates the famil iar, intelligent, and reliable impressions for the agent. In addition, the user’s listenable and understandable impressions of the news were also elevated. The effectiveness of the proposed method in interactive news reading by virtual agents was confirmed.
In this research, we report about a collaborative learning practice that scaffolds elementary-school children for deepening the children’s understanding about cyber-physical domain which requires relation-making between naïve-experiences and explanation model derived from the experiences. In the practice, we conducted “knowledge-construction jigsaw” which required the children to think about the question collaboratively: “What air-conditioners do for keeping rooms’ temperature nice for people?” using micro:bit and its programming. We collected pre- and post- answers of each child about the question, programs for operating micro:bit which were made by the children’s group, and discourse data with children’s actions (pointing PC’s screen, shaking micro:bit and so forth). The results from eight groups from twenty-four children suggested that all the groups completed to make expected program to answer the question. However, only four groups used actual micro:bit properly and the remaining four groups only used micro:bit-simulator which can be used in PCs. Comparative discourse analysis from these two categories of groups suggested that the children who were in actual micro:bit used groups more concerned physical conditions like checking room-temperature using an analog thermo-meter, and more hypothesized what will be shown on their micro:bit in relation to the room-temperature than the children in the no-use of micro:bit groups.
Language acquisition is supported by an ability called phonological awareness that allows children to become intentionally aware of units of phonology. It is known that erroneous pronunciation appears during the phonological awareness formation process. Towards developing supporting method of such developmental process, this research aims to examine the factors that induce and reduce such errors. To do so, we modeled phonological awareness using the cognitive architecture ACT-R and performed simulations that manipulated ACT-R parameters that correspond to both nature and nurture factors. As a result, it was confirmed that errors due to a lack of phonological awareness can be modeled with the innate memory retrieval mechanism. We also observed that such errors were reduced when learning factors were added to the model. However, we could not simulate this learning process. In the future, we will study the interaction task that enables learning to reduce phonological errors and contribute to the acquisition of phonological awareness.
Active learning has been gradually focused on in recent years and introduced in various education fields because it gives better learning experience and outcomes than conventional lecture style approach. In this active learning, the quality of interaction between students, teacher and students is critically important because it decides the outcomes. Therefore, it is necessary to evaluate the interaction properly at real time and improve it. In this study, we measured and analyzed the body movement of students in group work using the sensors on smartphone. Also, we evaluated the outcomes and subjective communication quality of group work. By analyzing these three relations, there are positive correlations between the amount of body movement and the outcomes, also the evaluation of communication of group work. These results show that it is possible to evaluate the learning activity of students at real time in group work by using smartphones.
This article discusses feasibility of detecting mislay of objects when they are left by users from their behavior. From questionnaires to the participants for the videos with mislay of objects, it has been found that time passed after the most recent manipulation of objects as well as number of objects left after being brought to the place serve as useful clues to detect mislay of objects. Since those clues can be obtained from videos using conventional methods of human pose estimation and object recognition, it is also shown that mislay of objects can be automatically detected from the videos.
In terms of communication robot, they can affect human communication through their behavior and expression; they are not just a machine but have the potential to make communication diverse. Most of the research on human-robot interaction so far focused on one-to-one interaction between humans and robots; communication robots often need to be fluent and perfect. On the other hand, we focus on the fact that our daily conversations are never perfectly conceived but contain incomplete statements in places due to the constraints of the cognitive resource. The incompleteness is often perceived as a negative phenomenon, but it also has positive power. The authors proposed an interaction design, focused on "forgetting" which is one of the incompleteness and developed the robot, Talking-Bones, that sometimes forgets words. Talking-Bones where the robot aims to communicate with humans in conversations recalling together. Talking-Bones shows encouraging behaviors by forgetting some words in a story and asking humans to help recalling the words together to continue telling the story. We conducted fieldwork to observe the interaction between children and Talking-Bones. These results illustrate how collaborative interactions are formed depending on the relationship of the children due to a forgetful robot.
Experimental tasks depicting a bullying scene are being studied to elucidate the causes of bullying and to verify methods for resolving bullying. Bullying involves not only victims, bullies, reinforcers, who are complicit in the bullying, and defenders, who mediate the bullying, but also bystanders, who ignore the bullying. Bystanders comprise the largest group involved in bullying and can play an important role in resolving bullying. However, in previous experimental tasks depicting a bullying scene, participants could not perceive bystanders’ behavior as different from the behaviors of the victim or bully. The present study aimed to contribute to the development of solutions to bullying by creating experimental tasks, including a measurable behavior representing bystanders. In this research, we introduced a new option of behavior representing bystanders in a “catch-ball task” that can express the reinforcing and defending behavior in a bullying scene. Results of our questionnaire survey showed that the newly implemented behavior representing bystanders was perceived as bystander’s behavior by participants. Moreover, according to the results of another questionnaire survey, the improved experimental tasks with one bystander were perceived to be closer to a bullying scene compared with previous catch-ball tasks.
Various tactile displays have been developed to present tactile graphics. Among them, electrostatic tactile displays are promising because of their low cost and flexibility. This study investigates the effectiveness of reading tactile graphics with multiple fingers on an electrostatic tactile display. For this purpose, we implemented an electrostatic tactile display using silver nanoparticle ink to allow multi-finger reading. Then, we conducted an experiment to compare one-finger and multi-finger reading with the implemented tactile display. The results showed that, in the multi-finger condition, the recognition rates were higher, and reading times were shorter, which indicates the effectiveness of multi-finger reading on the electrostatic tactile display. These results differ from other tactile displays, and is considered to be related to the size of the tactile graphics and factors specific to electrostatic tactile display. In addition, we observed the strategies for capturing a figure in each condition and examined the use of these strategies. By comparing these strategies with other related works, we consider that these strategies can be used as a reference for design guidelines regarding the presentation of tactile graphics with electrostatic tactile display.