Takahisa Uchida

In recent years, dialogue-support systems involving robots and AI have been spreading. However, the psychological factors that underpin people’s willingness to engage in dialogue remain unclear. We conducted an online vignette-based survey in which participants imagined having non-task-oriented conversations with three interlocutors (a robot, a voice-only AI agent, and a human; within-participant design). We examined how six psychological factors (trust, social presence, enjoyment, self-disclosure, anthropomorphism, and likeability) were associated with motivation to continue the dialogue, while treating negative attitudes toward robots (NARS) as a continuous moderator and controlling for age and gender. Linear mixed-effects modeling showed that enjoyment was the strongest predictor of dialogue motivation, followed by likeability and self-disclosure. Trust was also positively associated with motivation. NARS did not show a direct effect after controlling for these proximal factors, but it significantly moderated the effects of enjoyment and self-disclosure. Adjusted means differed slightly across interlocutors (AI > human), suggesting limited but detectable partner-type differences once perceptions were accounted for. We discuss design-relevant prioritization of interaction qualities while noting the constraints of vignette-based, self-reported intentions.

Evaluations in human-robot interaction (HRI) have traditionally focused on surface-level impressions such as likability and anthropomorphism, often neglecting the underlying relational structures that define mutual contributions and shared goals. Understanding these relational dynamics is crucial for fostering effective and sustainable human-robot collaborations. Here, we developed and validated a novel pictogram-based questionnaire grounded in social psychology to capture relational structures, conducting three scenario-based online surveys. We examined ideal and assumed relational structures in human-human and human-robot contexts across varying robot appearances, roles, and utterances. We distinguish between ideal relational structures (what users would like to establish; Surveys A and B) and assumed (expected) relational structures (what users believe would emerge under the described conditions; Survey C). Our results reveal that ideal relational structures vary by human relationship type, and that a robot’s communicative behavior significantly influences perceived relational structures, often more than physical appearance. These findings suggest that assessing relational perceptions can inform the design of robots better aligned with user expectations, enhancing long-term engagement and collaboration in HRI.

This study develops dialogue robots that encourage users to talk with the other directly to promote relationship building among them. It is said that friendly relationships with others have a positive impact on mental health and that face-to-face communication is important for building friendly relationships. However, building relationships with others is not easy, and many people find it difficult to find topics of conversation. While previous studies have supported relationship building using robots that share preferences, these studies are limited to users’ subjective evaluations, and it is not clear whether they can get people to talk directly to each other. This study proposes a dialogue robot based on identifying the common preferences with high information content and interests among the user and a third party. It also encourages direct conversation between them by creating a common question about the third party. To verify the effectiveness of the proposed method, we introduced the developed robots into an elementary school sixth-grade classroom for five days. The experimental results suggest that the proposed condition was more likely to increase the frequency of conversation among children in the pre-post period. The results also suggest that the proposed condition could be more likely to increase the frequency of conversation among children in the pre-week and pre-month periods. These results are valuable because they suggest the possibility that robots can support the construction of relationships among children in an actual elementary school setting. Future work can also verify the usefulness of the robots in various communities.

In android science, the greater the resemblance between an android and a human, the greater is the expectation of observers for its movements to appear natural; therefore, the "human-likeness" of android motion is important. Traditionally, studies on the human-likeness of motion have addressed two separate problems: generating motion trajectories and realizing the target trajectories on hardware. However, trajectory generation often relies on subjective evaluations and lacks quantitative metrics, whereas trajectory realization has limited reproducibility owing to the physical characteristics and control performance of the android. Therefore, treating these problems separately complicates the task of achieving human-like motion. This study proposes an approach that quantitatively addresses the human-likeness of such movements under the assumption that the target trajectory can be stably reproduced. As a control method, we adopted Active Disturbance Rejection Control (ADRC), which can suppress disturbances and produce stable responses even for complex nonlinear systems. We aim to express the human-likeness of movements realized under this control by using engineering metrics. Specifically, we computed the engineering metrics from motion data obtained by varying theADRCparameters and quantified the subjective perceptions of human-likeness based on pairwise comparison experiments. The analysis of the relationship between these two sets of measures confirmed a reasonable agreement between the ranking trends of the engineering metrics and the subjective evaluations. These results indicate that the proposed engineering metrics can quantitatively describe human-like motion independent of subjective assessments, and are effective as objective and unified indicators for android motion design.

Personalizing a dialogue system according to the user has been recognized to have various positive effects. Despite the significance of user values, concepts guiding choices and evaluations being recognized in communication, they have not been considered in personalized dialogue systems. Therefore, this study constructs a dialogue system that understands user values through conversation. Furthermore, the impact of understanding values on the interactions between dialogue systems and users is examined. The method is organized with a user model of preferences and values based on the established means-end chain model. We used a large language model (LLM) to estimate values based on the users’ preferences and the reasons they prefer them. Furthermore, an infinite relational model (IRM) estimates the relationships between multiple elements within the user model. The experiments show that the proposed method could estimate user values and enhance animacy and perceived intelligence in users’ impressions of an android robot, prompting new insights into users’ own values. The perception of the robot contributes to improved-quality interactions, and new insights into values facilitate a deeper self-understanding of users. This achievement, demonstrating the effects of using values for interaction, can provide valuable insights into human-robot interaction.

This study employs Conversation Analysis to create a recursive model that improves the quality of human-robot interaction. Our research goal is to create a dialogue robot that offers pleasant experiences for users, so they are willing to engage in repeated interactions in daily lives. While there has been dramatic progress in the performance of dialogue robots, there has been less attention to the importance of users’ interactional experience compared to the “specs” of the dialogue system. Employing Goffmanian insights and using research in Conversation Analysis (CA), the present study develops a dialogue closing system to exit the interaction. We then experimentally verified that the robot with the dialogue closing system performs better in the user’s perception of the robot (i.e. likeability, politeness, and dialogue satisfaction) than the control group. Further, by analyzing the dialogue between the human and the robot through CA, we propose to build a recursive, reflective model to improve the dialogue model design. A constructive approach urges us to reproduce complicated social phenomena in human-robot interaction so that we can investigate the underlying cognitive mechanisms of humans and create robots that can convey human-like cognition functions and coexist with humans. Taking such a constructive approach, we posit that our recursive model for dialogue systems that uses CA insights and then qualitatively analyzes conversational data can enhance the quality of dialogue systems because the model elucidates which properties of a conversation humans need to experience a conversational robot as human-like. Our study suggests that interactional morality - particularly conversational closings - is one property of human interactions that humans likely require social robots to adhere to.

With the age of diversification, it’s important to nurture the qualities of children to enable them to create their future society. To achieve this, it’s crucial for children to broaden and deepen their views through dialogue with others. In the field of human-agent interaction, introducing robots into educational settings has shown positive educational effects. In this study, we developed an autonomous dialogue robot that delved into the children’s ideas and prompted them to think about aspects they had not yet considered, to broaden and deepen their viewpoints. We introduced this developed robot into a 6th-grade science class and conducted an experiment. The results based on subjective evaluations by students through questionnaires and objective evaluations of the students’ reaction papers by third parties, suggest that discussions with the robot can broaden and deepen the viewpoints of students. This study highlights the potential of robots to support the important educational goal.

The goal of this study is to maintain dialogue motivation between users through the mediation of a dialogue robot. Previous studies proposed artificial agents that present dialogue topics to facilitate relationship-building among individuals for a short period. However, sustaining dialogue motivation between users using such agents has not been investigated. This study proposes an algorithm for presenting topics about other people to sustain dialogue motivation between users for certain periods. Specifically, we designed a dialogue robot that discusses topics of common preference, emphasizing high information content between users. To validate our approach, we applied the proposed algorithm to a dialogue robot and conducted experiments involving university and graduate students belonging to the same community. The results confirmed that our proposed method prevented them from decreasing their engagement in dialogue even after one month of their interaction with the robot. Additionally, topics with high information content were more likely to be remembered by users. In other words, these findings indicate that when a robot introduces a topic with high information content, participants might perceive it as unusual. This perception prompts them to retain the topic with clarity; as a consequence, it might enhance the participants’ willingness to engage in dialogue with each other. Based on the findings of this research, using highly informative topics in dialogue can be an effective strategy for cultivating long-term relationships among individuals. This also insight emphasizes the potential role of robots in facilitating human relationship-building. Future studies need to examine the effectiveness of the proposed method with various communities.