Geminoid Studies

Frontmatter

Chapter 1. Development of an Android System Integrated with Sensor Networks

Abstract

To develop a robot that has a humanlike presence, the robot must be given very humanlike appearance and behavior, and a sense of perception that enables it to communicate with humans. We have developed an android robot called “Repliee Q2” that closely resembles human beings; however, sensors mounted on its body are not sufficient to allow humanlike communication with respect to factors such as the sensing range and spatial resolution. To overcome this issue, we endowed the environment surrounding the android with perceptive capabilities by embedding it with a variety of sensors. This sensor network provides the android with humanlike perception by constantly and extensively monitoring human activities in a less obvious manner. This paper reports on an android system that is integrated with a sensor network system embedded in the environment. A human–android interaction experiment shows that the integrated system provides relatively humanlike interaction.

Takenobu Chikaraishi, Takashi Minato, Hiroshi Ishiguro

Chapter 2. Building Artificial Humans to Understand Humans

Abstract

If we could build an android as a very humanlike robot, how would we, humans, distinguish a real human from the android? The answer to this question is not simple. In human–android interactions, we cannot see the internal mechanism of the android, and thus, we may simply believe that it is a human. This means that humans can be defined in two ways: by their organic mechanism and by their appearance. Further, the current rapid progress in the development of artificial organs makes this distinction confusing. The approach discussed in this paper is to create artificial humans based on humanlike appearances. The developed artificial humans, an android and a geminoid, can be used to understand humans through psychological and cognitive tests. We call this new approach to understanding humans “Android Science.”

Hiroshi Ishiguro, Shuichi Nishio

Chapter 3. Androids as a Telecommunication Medium with a Humanlike Presence

Abstract

In this study, we realize human telepresence by developing a remote-controlled android system called Geminoid HI-1. Experimental results confirm that participants feel a stronger presence of the operator when he talks through the android than when he appears on a video monitor in a video conference system. In addition, participants talk with the robot naturally and evaluate its humanlike-ness as equal to a man on a video monitor. We also discuss a remote-controlled system for telepresence that uses a humanlike android robot as a new telecommunication medium.

Daisuke Sakamoto, Takayuki Kanda, Tetsuo Ono, Hiroshi Ishiguro, Norihiro Hagita

Chapter 4. Generating Natural Motion in an Android by Mapping Human Motion

Abstract

One of the main aims of humanoid robotics is to develop robots that are capable of interacting naturally with people. However, to understand the essence of human interaction, it is crucial to investigate the contribution of behavior and appearance. Our group’s research explores these relationships by developing androids that closely resemble human beings in both aspects. If humanlike appearance causes us to evaluate an android’s behavior from a human standard, we are more likely to be cognizant of deviations from human norms. Therefore, the android’s motions must closely match human performance to avoid looking strange, including such autonomic responses as the shoulder movements involved in breathing. This paper proposes a method to implement motions that look human by mapping their three-dimensional appearance from a human performer to the android and then evaluating the verisimilitude of the visible motions using a motion capture system. Previous research has focused on copying and moving joint angles from a person to a robot. Our approach has several advantages: (1) in an android robot with many degrees of freedom and kinematics that differ from that of a human being, it is difficult to calculate which joint angles would make the robot’s posture appear similar to the human performer; and (2) the motion that we perceive is at the robot’s surface, not necessarily at its joints, which are often hidden from view.

Daisuke Matsui, Takashi Minato, Karl F. MacDorman, Hiroshi Ishiguro

Chapter 5. Formant-Based Lip Motion Generation and Evaluation in Humanoid Robots

Abstract

Generating natural motion in robots is important for improving human–robot interaction. We have developed a teleoperation system in which the lip motion of a remote humanoid robot is automatically controlled by the operator’s voice. In the present work, we introduce an improved version of our proposed speech-driven lip motion generation method, where lip height and width degrees are estimated based on vowel formant information. The method requires the calibration of only one parameter for speaker normalization. Lip height control is evaluated in two types of humanoid robots (Telenoid-R2 and Geminoid-F). Subjective evaluations indicate that the proposed audio-based method can generate lip motion with superior naturalness to vision-based and motion capture-based approaches. Partial lip width control is shown to improve lip motion naturalness in Geminoid-F, which also has an actuator for stretching the lip corners. Issues regarding online real-time processing are also discussed.

Carlos T. Ishi, Chaoran Liu, Hiroshi Ishiguro, Norihiro Hagita

Chapter 6. Analysis of Head Motions and Speech, and Head Motion Control in an Android Robot

Abstract

With the aim of automatically generating head motions during speech utterances, analyses are conducted to verify the relations between head motions and linguistic and paralinguistic information carried by speech utterances. Motion-captured data are recorded during natural dialogue, and the rotation angles are estimated from the head marker data. Analysis results show that nods frequently occur during speech utterances, not only for expressing specific dialogue acts such as agreement and affirmation, but also to indicate syntactic or semantic units, which appear at the last syllable of the phrases, in strong phrase boundaries. The dependence on linguistic, prosodic and voice quality information of other head motions, including shakes and tilts, is also analyzed, and the potential for using this to automatically generate head motions is discussed. Intra-speaker variability and inter-speaker variability on the relations between head motion and dialogue acts are also analyzed. Finally, a method for controlling the head actuators of an android based on the rotation angles is proposed, and the mapping from human head motions is evaluated.

Carlos Toshinori Ishi, Hiroshi Ishiguro, Norihiro Hagita

Chapter 7. Generation of Head Motion During Dialogue Speech, and Evaluation in Humanoid Robots

Abstract

Head motion occurs naturally and in synchrony with speech during human dialogue communication and may carry paralinguistic information such as intentions, attitudes, and emotions. Therefore, natural-looking head motion by a robot is important for smooth human–robot interaction. Based on rules inferred from analyses of the relationship between head motion and dialogue acts, we proposed a model for generating nodding and head tilting and evaluated for different types of humanoid robot. Analysis of subjective scores showed that the proposed model including head tilting and nodding can generate head motion with increased naturalness compared to nodding only or directly mapping people’s original motions without gaze information. We also found that an upward motion of the face can be used by robots that do not have a mouth in order to provide the appearance that an utterance is taking place. Finally, we conducted an experiment in which participants act as visitors to an information desk attended by robots. Evaluation results indicated that our model is equally effective as directly mapping people’s original motions with gaze information in terms of perceived naturalness.

Carlos T. Ishi, Chaoran Liu, Hiroshi Ishiguro

Chapter 8. Uncanny Valley of Androids and the Lateral Inhibition Hypothesis

Abstract

From the viewpoint of designing a robot for communication, it is important to avoid the ‘uncanny valley,’ although this is an essential phenomenon for discovering the principles relevant to establishing and supporting social interaction between humans and robots. Studying the uncanny valley allows us to explore the boundary of humanlike-ness. We have empirically and experimentally obtained evidence for the uncanny valley effect, which has thus far only been hypothesized through the development of androids that closely resemble human beings. We have also obtained experimental evidence to suggest that the uncanny valley varies owing to the development of individuals. We refer to this variable uncanny valley as the age-dependent uncanny valley. We assume that the uncanny valley is induced by a lateral inhibition effect, which is the same mechanism observed in sensory cells, and is referred to herein as the lateral inhibition hypothesis of the uncanny valley. The present paper presents evidence concerning the uncanny valley and describes the likelihood of the present hypothesis.

Michihiro Shimada, Takashi Minato, Shoji Itakura, Hiroshi Ishiguro

Chapter 9. Evaluation of Robot Appearance Using a Brain Science Technique

Abstract

We evaluate the humanlike-ness of humanoid robots using electroencephalography (EEG). As the activity of the human mirror-neuron system (MNS) is believed to reflect the humanlike-ness of observed agents, we compare the MNS activity of 17 participants while observing certain actions performed by a human, an extremely humanlike android, and a machine-like humanoid. We find the MNS to be significantly activated only when the participants observe actions performed by the human. Despite the participants’ rating of the android appearance as more humanlike than that of the robot, the MNS activity corresponding to each of the three agents does not differ. These findings suggest that appearance does not crucially affect MNS activity, and that factors such as motion should be targeted for improving the humanlike-ness of humanoid robots.

Goh Matsuda, Kazuo Hiraki, Hiroshi Ishiguro

Chapter 10. Persistence of the Uncanny Valley

Abstract

In recent years, the uncanny valley theory has been heavily investigated by researchers from various fields. However, the videos and images used in these studies do not permit any human interaction with the uncanny objects. Therefore, in the field of human–robot interaction, it is still unclear what impact, if any, an uncanny-looking robot will have in the context of an interaction. In this paper, we describe an exploratory empirical study using a live interaction paradigm that involves repeated interactions with robots that differ in embodiment and their attitude toward humans. We find that both components of uncanniness investigated here (likeability and eeriness) can be affected by an interaction with a robot. The likeability of a robot is mainly affected by its attitude, and this effect is especially prominent for a machinelike robot. Merely repeating interactions is sufficient to reduce the degree of eeriness, irrespective of a robot’s embodiment. As a result, we urge other researchers to investigate the uncanny valley theory in studies that involve actual human–robot interactions in order to fully understand the changing nature of this phenomenon.

Jakub A. Złotowski, Hidenobu Sumioka, Shuichi Nishio, Dylan F. Glas, Christoph Bartneck, Hiroshi Ishiguro

Chapter 11. Can a Teleoperated Android Represent Personal Presence?—A Case Study with Children

Abstract

Our purpose is to investigate the key elements for representing personal presence, which is the sense of being with a certain individual. A case study is reported in which children performed daily conversational tasks with a geminoid, a teleoperated android robot that resembles a living individual. Different responses to the geminoid and the original person are examined, with a special focus on the case where the target child was the daughter of the geminoid source. Our results show that children gradually adapt to conversation with the geminoid, but the operator’s personal presence is not represented completely. Further research topics on the adaptation process to androids and the key elements of personal presence are discussed.

Shuichi Nishio, Hiroshi Ishiguro, Norihiro Hagita

Chapter 12. Cues that Trigger Social Transmission of Disinhibition in Young Children

Abstract

Previous studies have shown that observing the actions of a human model, but not those of a robot, can induce perseverative behaviors in young children, suggesting that children’s socio-cognitive abilities may lead to perseverative errors (“social transmission of disinhibition”). This study investigates how the social transmission of disinhibition occurs. Specifically, the authors examine whether a robot with human appearance (an android) triggers perseveration in young children and compare the effects with those induced by a human model. The results reveal that the android induces the social transmission of disinhibition. Additionally, children are more likely to be affected by the human model than by the android. The results suggest that behavioral cues (biological movement) may be important for the social transmission of disinhibition.

Yusuke Moriguchi, Takashi Minato, Hiroshi Ishiguro, Ikuko Shinohara, Shoji Itakura

Chapter 13. Effects of Observing Eye Contact Between a Robot and Another Person

Abstract

One of the common requirements for a communication robot is to be accepted by humans. Previous work has examined the effects of nonverbal factors on people’s perceptions of robots for such a purpose, but always with a focus on dyadic human–robot interaction; in real human society, however, triadic interaction also plays an important role and should be considered. This paper explores a potential merit offered by the latter form of interaction; specifically, how one form of nonverbal interaction occurring between a robot and humans, eye contact, can be utilized to make the robot appear more acceptable to humans. Experiments are conducted with groups of two humans and an android. One of the humans, the “subject,” is asked to communicate with a second person, the “confederate,” who knows the purpose of the experiment; the confederate’s role is to gaze in such a way that the subject either observes or does not observe eye contact between the confederate and the android. A post-interaction questionnaire reveals that subjects’ impressions toward the robot are influenced by eye contact between the confederate and the robot. Finally, the consistency of the experimental results is discussed in terms of Heider’s balance theory, and future extensions of this research are proposed.

Michihiro Shimada, Yuichiro Yoshikawa, Mana Asada, Naoki Saiwaki, Hiroshi Ishiguro

Chapter 14. Can an Android Persuade You?

Abstract

The first robotic copies of real humans have become available. They enable their users to be physically present in multiple locations simultaneously. This study investigates the influence that the embodiment of an agent has on its persuasiveness and its perceived personality. Is a robotic copy as persuasive as its human counterpart? Does it have the same personality? We performed an experiment in which the embodiment of the agent was the independent variable and the persuasiveness and perceived personality were the dependent measurements. The persuasive agent advertised a Bluetooth headset. The results show that an android is perceived as being as persuasive as a real human or a video recording of a real human. The personality of the participant had a considerable influence on the measurements. Participants who were more open to new experiences rated the persuasive agent lower on agreeableness and extroversion. They were also more willing to spend money on the advertised product.

Kohei Ogawa, Christoph Bartneck, Daisuke Sakamoto, Takayuki Kanda, Tetsuo Ono, Hiroshi Ishiguro

Chapter 15. Attitude Change Induced by Different Appearances of Interaction Agents

Abstract

Human–robot interaction studies have thus far been limited to simple tasks, such as route guidance or playing simple games. However, with the advance in robotic technologies, the stage has been reached where the requirements for highly complicated tasks, such as conducting humanlike conversations, should be explored. When robots start to play advanced roles in our lives, such as in health care, attributes such as the trust and reliance of the person and the persuasiveness of the robot also become important. In this study, we examined the effect of the appearance of robots on people’s attitudes toward them. Past studies have shown that the appearance of robots is one of the elements that influence people’s behavior. However, the effect of the robots appearance on a person when conducting serious conversations that require high-level activity remains unknown. Participants were asked to hold a discussion with teleoperated robots having various appearances, such as an android with a high similarity to a human and a humanoid robot having humanlike body parts. Through the discussion, the teleoperator attempted to persuade the participants. We examined the effect of the robots appearance on their persuasiveness, as well as on people’s behavior and their impression of the robots. A possible contribution to machine consciousness research is also discussed.

Shuichi Nishio, Hiroshi Ishiguro

Chapter 16. Do Robot Appearance and Speech Affect People’s Attitude? Evaluation Through the Ultimatum Game

Abstract

In this study, we examined the factors that influence humans recognition of robots as social beings. Participants took part in sessions of the Ultimatum Game, a procedure frequently used in the fields of both economics and social psychology for examining peoples’ attitudes toward others. In the experiment, several agents having different appearances were used, and speech stimuli, which were expected to induce a mentalizing effect on the participants’ attitude toward the agents, were utilized. The results show that, while appearance per se did not elicit a significant difference in the participants attitudes, the mentalizing stimuli affected their attitudes in different fashions, depending on the robots’ appearance. These results showed that elements such as a simple conversation with an agent and the agents’ appearance are important factors that cause people to treat robots in a more humanlike fashion and as social beings.

Shuichi Nishio, Kohei Ogawa, Yasuhiro Kanakogi, Shoji Itakura, Hiroshi Ishiguro

Chapter 17. Isolation of Physical Traits and Conversational Content for Personality Design

Abstract

In this paper, we propose the “Doppel teleoperation system,” which isolates several physical traits of a speaker, for investigating the manner in which personal information is conveyed to others during conversation. An underlying problem related to designing a personality for a social robot is that the manner in which humans judge the personalities of conversation partners remains unclear. The Doppel system allows each of the communication channels to be transferred, so that the channel can be chosen in its original form or in that generated by the system. For example, voice and body motions can be replaced by the Doppel system, while the speech content is preserved. This allows the individual effects of the physical traits of the speaker and the content of the speaker’s speech on the identification of personality to be identified. This selectivity of personal traits provides a useful approach for investigating which information conveys our personality through conversation. To show the potential of our system, we experimentally tested the extent to which the conversation content conveys the personality of speakers to interlocutors when their physical traits are eliminated. The preliminary results show that, although interlocutors experience difficulty identifying speakers using only conversational content, when their acquaintances are the speakers, they can recognize them. We indicate several potential physical traits that convey personality.

Hidenobu Sumioka, Shuichi Nishio, Erina Okamoto, Hiroshi Ishiguro

Chapter 18. Body Ownership Transfer to a Teleoperated Android

Abstract

Teleoperators of android robots occasionally feel as if the robotic bodies are extensions of their own bodies. When others touch the android that they are teleoperating, even without tactile feedback, some operators feel as if they themselves have been touched. In the past, a similar phenomenon named the “Rubber Hand Illusion” was studied because it reflects the three-way interaction between vision, touch, and proprioception. In this study, we examined whether a similar three-way interaction occurs when the tactile sensation is replaced with android robot teleoperation. The results show that when the operator and the android motions are synchronized, operators feel as if their sense of body ownership is transferred to the android robot.

Shuichi Nishio, Tetsuya Watanabe, Kohei Ogawa, Hiroshi Ishiguro

Chapter 19. Effect of Perspective Change on Body Ownership Transfer

Abstract

We previously investigated body ownership transfer to a teleoperated android body caused by motion synchronization between the robot and its operator. Although visual feedback is the only information provided from the robot side, as a result of body ownership transfer, some operators feel as if they are touched when the robot’s body is touched. This illusion can help operators transfer their presence to the robotic body during teleoperation. By enhancing this phenomenon, we can improve the communication interface and the quality of the interaction between the operator and interlocutor. In this study, we examined the effect of a change in the operator’s perspective on the body ownership transfer during teleoperation. According to the results of past studies on the rubber hand illusion (RHI), we hypothesized that a perspective change would suppress the body owner transfer. Our results, however, showed that under any perspective condition, the participants felt the body ownership transfer. This shows that its generation process differs for teleoperated androids and the RHI.

Kohei Ogawa, Koichi Taura, Shuichi Nishio, Hiroshi Ishiguro

Chapter 20. Body Ownership Transfer by Social Interaction

Abstract

Body ownership transfer (BOT) comprises the illusion that we feel external objects as parts of our own body, which occurs when teleoperating android robots. In past studies, we investigated the conditions under which this illusion occurs. However, these studies were conducted using only simple operation tasks, such as moving only the robot’s hand. Does this illusion occur during more complex tasks, such as conducting a conversation? What kind of conversation setting is required to invoke this illusion? In this study, we examined the manner in which factors in social interaction affect the occurrence of BOT. Participants conversed using the teleoperated robot under different conditions and teleoperation settings. The results revealed that BOT does occur during the task of conducting a conversation, and that the conversation partner’s presence and appropriate responses are necessary to enhance BOT.

Shuichi Nishio, Koichi Taura, Hidenobu Sumioka, Hiroshi Ishiguro

Chapter 21. Exploring Minimal Requirement for Body Ownership Transfer by Brain–Computer Interface

Abstract

Operators of a pair of robotic hands report ownership of those hands when they hold an image in their mind of a grasp motion and watch the robot perform it. We present a novel body ownership illusion that is induced by merely watching and controlling a robot’s motions through a brain–machine interface. In past studies, body ownership illusions were induced by the correlation of sensory inputs, such as vision, touch, and proprioception. However, in the presented illusion none of these sensations was integrated, except vision. Our results show that the interaction between the motor commands and visual feedback of an intended motion is sufficient to evoke the illusion that non-body limbs are incorporated into a person’s own body. In particular, this work discusses the role of proprioceptive information in the mechanism of agency-driven illusions. We believe that our findings can contribute to the improvement of tele-presence systems in which operators perceive tele-operated robots as themselves.

Maryam Alimardani, Shuichi Nishio, Hiroshi Ishiguro

Chapter 22. Regulating Emotion with Body Ownership Transfer

Abstract

In this study, we experimentally examined whether changes in the facial expressions of teleoperated androids can affect and regulate their operators’ emotion, based on the facial feedback theory of emotion and the phenomenon of body ownership transfer to the robot. Twenty-six Japanese participants conversed with an experimenter through a robot in a situation where the participants were induced to feel anger, and during the conversation, the android’s facial expression was changed according to a pre-programmed scheme. The results showed that facial feedback from the android did occur. Moreover, a comparison of the results of two groups of participants, one of which operated the robot and the second did not, showed that this facial feedback from the android robot occurred only when the participants operated the robot, and that when an operator could effectively operate the robot, his/her emotional states were more affected by the facial expression change of the robot.

Shuichi Nishio, Koichi Taura, Hidenobu Sumioka, Hiroshi Ishiguro

Chapter 23. Adjusting Brain Activity with Body Ownership Transfer

Abstract

Feedback design is an important issue in motor imagery brain–computer interface (BCI) systems. However, extant research has not reported on the manner in which feedback presentation optimizes coadaptation between a human brain and motor imagery BCI systems. This study assesses the effect of realistic visual feedback on user BCI-performance and motor imagery skills. A previous study developed a teleoperation system for a pair of humanlike robotic hands and showed that the BCI control of the hands in conjunction with first-person perspective visual feedback of movements arouses a sense of embodiment in the operators. In the first stage of this study, the results indicated that the intensity of the ownership illusion was associated with feedback presentation and subject performance during BCI motion control. The second stage investigated the effect of positive and negative feedback bias on BCI-performance of subjects and motor imagery skills. The subject-specific classifier that was set up at the beginning of the experiment did not detect any significant changes in the online performance of subjects, and the evaluation of brain activity patterns revealed that the subject’s self-regulation of motor imagery features improved due to a positive feedback bias and the potential occurrence of ownership illusion. The findings suggest that the manipulation of feedback can generally play an important role with respect to training protocols for BCIs in the optimization of the subject’s motor imagery skills.

Maryam Alimardani, Shuichi Nishio, Hiroshi Ishiguro

Chapter 24. At the Café—Exploration and Analysis of People’s Nonverbal Behavior Toward an Android

Abstract

Current studies investigating natural human–robot interaction (HRI) in the field concentrate on the analysis of automatically assessed data (e.g., interaction times). What is missing to date is a more qualitative approach to investigate the natural and individual behavior of people in HRI in detail. In a quasi-experimental observational field study, we investigated how people react to an android robot in a natural environment according to the behavior it exhibits. We present data on unscripted interactions between humans and the android robot “Geminoid HI-1” in an Austrian public café and subsequent interviews. Data related to the participants’ nonverbal behavior (e.g., attention paid to the robot and proximity) were analyzed. The results show that participants’ behavior toward the android robot, as well as their interview answers, was influenced by the behavior the robot exhibited (e.g., eye contact). In addition, huge inter-individual differences existed in the participants’ behavior. Implications for HRI research are discussed.

Astrid M. von der Pütten, Nicole C. Krämer, Christian Becker-Asano, Kohei Ogawa, Shuichi Nishio, Hiroshi Ishiguro

Chapter 25. At the Café—from an Object to a Subject

Abstract

What are the characteristics that make an object appear to be a social entity? Is sociality limited to human beings? This article addresses the borders of sociality and the animation characteristics with which a physical object (here, an android robot) needs to be endowed so that it appears to be a living being. The transition of sociality is attributed during interactive encounters. We introduce the implications of an ethnomethodological analysis to show the characteristics of the transitions in the social attribution of an android robot, which is treated and perceived as gradually shifting from an object to a social entity. These characteristics should (a) fill the gap in current anthropological and sociological research, addressing the limits and characteristics of social entities and (b) contribute to the discussion of the specific characteristics of human–android interaction as compared to human–human interaction.

Ilona Straub, Shuichi Nishio, Hiroshi Ishiguro

Chapter 26. At the Hospital

Abstract

Recently, many humanoid robots have been developed and actively investigated all over the world in order to realize partner robots which coexist in the human environment. For such robots, high communication ability is essential in order to naturally interact with human. However, even with the state-of-the-art interaction technology, it is still difficult for human to interact with humanoids without conscious efforts. In this chapter, we apply the android robot, which has a quite similar appearance to a human, to a bystander in human-human communication. The android is not explicitly involved in the conversation, however makes small reactions to the behaviors of the humans, and the psychological effects on human subjects are investigated. Through the experiments, it is shown that mimicry behavior of the subject by the android is quite effective to harmonize the human-human communication.

Eri Takano, Yoshio Matsumoto, Yutaka Nakamura, Hiroshi Ishiguro, Kazuomi Sugamoto

Chapter 27. At the Department Store—Can Androids Be a Social Entity in the Real World?

Abstract

In this paper, we discuss an autonomous android robot that is recognized as a social entity by observers in the real world. We conducted field experiments to investigate the type of function with which an android should be provided so that it is can be recognized as a social and humanlike entity by observers. In the field experiment, the android conversed with multiple visitors through several touch displays. The results show that the visitors evaluated the android as humanlike, although this type of interaction differs from normal human–human interaction. Moreover, the results of the experiment suggest that the android exerts a social influence-advertisement effect on visitors.

Miki Watanabe, Kohei Ogawa, Hiroshi Ishiguro

Chapter 28. At the Department Store—Can Androids Be Salespeople in the Real World?

Abstract

The roles of robots in the real world have become more diverse according to the bodily properties with which they are endowed. In this study, we aimed to determine the roles that androids, the bodily properties of which resemble those of humans, could serve in the real world. Selling and purchasing are common human activities. Therefore, we propose the use of an android as a salesperson utilizing cognitive and affective strategies that exploit the advantages of online- and counter-selling methods. We conducted a field study to investigate whether androids can sell goods in a department store. The results show that the sales strategies were effective and that over 10 days the android sold 43 sweaters that cost approximately $100 each. These results provide important knowledge for determining how androids may fill new roles and communicate with humans in the real world.

Miki Watanabe, Kohei Ogawa, Hiroshi Ishiguro

Chapter 29. At the Theater—Designing Robot Behavior in Conversations Based on Contemporary Colloquial Theatre Theory

Abstract

The design of humanlike behavior for a robot that interacts with humans remains a central issue in the human–robot interaction (HRI) field because of humans sensitivity to humanlike objects. This issue is very challenging, because humanlikeness is an important factor in designing better interactions, since an imperfect design can easily cause negative impressions, as reflected by the uncanny valley phenomenon. This paper addresses this issue using a novel approach that utilizes implicit know-how for performing on stage dedicated to the stage representation of human beings. Contemporary colloquial theatre theory (CCTT), which is a theory applied in a method of directing plays, is appropriate for this purpose, since its reality-oriented instructions are directly applicable to improving robot behavior. In this paper, we report a case study involving the performance of a play in which both humans and a robot played roles. The play in our study was evaluated by the audiences of public performances in Japan. We also report a detailed analysis of HRI or human–human–robot interaction in comparable short plays. Our analysis implies that the robots utterances and motion timings should be tuned according to the situation. In future work, a motion capture system will be applied to obtain more precise data and more useful knowledge.

Kohei Ogawa, Takenobu Chikaraishi, Yuichiro Yoshikawa, Oriza Hirata, Hiroshi Ishiguro

Chapter 30. At the Theater—Possibilities of Androids as Poetry-Reciting Agents

Abstract

In recent years, research on very humanlike androids has increased, in general investigating the following: (1) the manner in which people treat these very humanlike androids and (2) whether it is possible to replace existing communication media, such as telephones or TV conference systems, with androids as a communication medium. We found that androids have advantages over humans in specific contexts. For example, in a collaborative theatrical project between artists and androids, audiences were impressed by the androids that read poetry. We therefore experimentally compared androids and humans in a poetry-reciting context by conducting an experiment to illustrate the influence of an android who recited poetry. Participants listened to poetry that was read by three poetry-reciting agents: an android, a human model on which the android was based, and a box. The experimental results show that the participants scored their enjoyment of the poetry highest under the android condition, indicating that the android has an advantage for communicating the meaning of poetry.

Kohei Ogawa, Hiroshi Ishiguro