Making it easier to design interactions between agents and humans is essential for realizing multi-agent simulations of social phenomena such as group dynamics. To realize large-scale social simulations, we have developed the scenario description languages Q and IPC (Interaction Pattern Card); they enable experts in the application domain (often not computing professionals) to easily create complex scenarios. We have also established a four-step process for creating scenarios: 1) defining a vocabulary, 2) describing scenarios, 3) extracting interaction patterns, and 4) integrating real and virtual experiments. In order to validate the scenario description languages and the four-step process, we ran a series of evacuation simulations based on the proposed languages and process. We successfully double-check the result of the previous controlled experiment done in a real environment.

People communicating through machine translators cannot tell what the purpose of their communication is or what other people are thinking because of the poor quality of translation services. If they are able to share their understanding within a "common ground" like a communicative or behavioral protocol, they can overcome their difficulties in communication, and we can improve information systems to help them improve mutual understanding. We designed a multilingual participatory gaming simulation, and conducted multilingual gaming experiments with Japanese and Korean participants. We extracted the protocol for conversation with mistranslations from the game logs and designed an agent to support conversation. Then, Japanese and Chinese played it and we observed and analyzed the behaviors of agents and the interaction between players and agents. Consequently, we obtained two main sets of results: (1) an agent function that notified players of the time that had elapsed since the conversation had broken down effectively speeded up their negotiations and achieved more active communications. (2) Tagging by participants was difficult and ineffective in leading to specific protocols and conversations when mistranslations occurred.

No abstract

A multilingual gaming simulation is suggested as the basis of an experiment by which we discuss complex problems such as global civics or environmental problems. Assigning tags is a very important way to analyze this experiment. However, assigning tags has a cost that is too large for analysts. In this study, we suggest and introduce a method for participants to voluntarily self-tag, focusing on gamification and crowdsourcing. As a result, we verified that it is possible for analysts to reduce the tagging cost when analyzing and to analyze results more accurately.

No abstract