This study investigates the influence of agents’ behavioral characteristics on human decision-making and the manner by which humans evaluate and treat agents, specifically in situations where they interact with agents adopting different strategies. Based on the ultimatum game with egocentric, exocentric, and adaptive agents, the findings suggest that inequity aversion is most likely to be expressed when the agent adopts an egocentric strategy. Human decision-making in interactions with agents is shaped not only by a preference for fairness but also by a self-serving tendency to prioritize one’s own benefits, provided that the opponent accepts the proposal. Furthermore, when the distribution is more favorable to the participants, they tend to evaluate the agent’s intelligence more negatively.

Using the Miracle Question method, this study comparatively evaluated how older adults and university students were affected by interacting with an active listening counseling robot for three weeks. The research focused on two aspects: (1) evaluating changes in perceptions of the robot and (2) how conversational content influenced these impressions. The results showed that older adults initially rated the robot higher for Anthropomorphism, with increased ratings for Perceived Intelligence and Anthropomorphism by Week 3. Both groups rated Perceived Safety the highest, linked to the prevalence of positive remarks during the counseling session conversations. These findings highlight the potential of positive dialogue strategies for counseling robots to engender trust and emotional safety across these age groups.

This paper presents collaborative concept-mapping tutor (CoCot ver.2), a collaborative learning support system that integrates concept maps with a conversational agent. CoCot ver.2 features two agents: a teacher agent and a student agent. The teacher agent acts as a human instructor, engaging in conversations with learners, aiding their metacognition, and summarizing the discussion content. The student agent learns from the learners' concept map creation and generates their own concept map knowledge. These agents are developed using (1) a cognitive architecture (ACT-R) for knowledge generation for the agents' concept map and (2) GPT 3.5 for part of the language processing for agent-based feedback.

In this report, we investigate the factor of social presence of a robot by using an actual robot and comparing it with a CG robot studied in our previous study. A laboratory experiment is conducted using a simple jury decision-making task, where participants play the role of a jury and make decisions regarding the length of the sentence for a particular crime. During the task, a robot with expert knowledge provides suggestions regarding the length of the sentence based on other similar cases. Results show that participants who engaged with an actual robot showed higher conformity with the suggested length of a sentence compared to the participants who engaged with a CG robot presented through a computer monitor. This study shows results that are consistent with those of previous studies in that interacting with physically aware robots is more engaging and also shows its effects on decision-making in a court.

Studies in collaborative learning have shown that explanation activities drawing on diverse perspectives facilitate deeper understanding and metacognition. However, it is difficult to develop an explicit understanding of others’ perspectives and knowledge through communication in a computer-mediated environment. The present study investigated the use of a visually shared concept map interface, expected to facilitate dyadic awareness of different perspectives and thus improve learning performance during explanation activities. In this study, each dyad built a concept map about a key technical term in psychology, and generated explanations of the term and generated explanations of the term on mutually accessible concept maps. We predicted that learners would be able to (1) gain deeper knowledge through the shared explanations; and (2) explain the key term from different perspectives or knowledge sets. Twenty-six university students participated in this experiment, and we assessed their performance through free recall tests before and after they used the concept mapping tool. Our findings showed that learners were able to (1) gain learning performance and (2) explain a concept based on different perspectives. We discuss the implications of our findings and suggest directions for further research on the development of learning support systems.