本研究旨在設計並建構議題導向、虛實整合之數位學習系統〈烏托邦〉。系統以複雜社會議題為核心，設置六個具有不同資源條件與發展特質的虛擬國家，功能包含世界與國家參數即時呈現、宣言發布、交易協商、行動卡經營，以及由〈烏托邦MGM〉AI遊戲助手評估各國決策並觸發突發事件。活動以虛實整合方式實施，學習者於實體討論中進行協商，並透過數位系統完成行動選擇與歷程記錄。 本研究對象為臺灣北部某國小四年級學生共27名，以4至5人為一組，扮演資源國、市場國、能源國、消費國、農業國及發展國之國家代表，完成兩回合的事件討論、宣言發布、交易談判與行動卡經營。研究以質性為主、量化問卷為輔，透過行為編碼表分析宣言內容與交易紀錄，以關係定義編碼表判定國家互動關係型態，並以描述性統計呈現滿意度問卷結果。 研究結果顯示，學習者面對事件時以多重目的為主要行動模式，同時兼顧解決事件、經營數值與交易談判；面對突發事件則以解決事件為單一目的，顯示全球性情境促使學習者優先聚焦問題處理。在互動關係方面，各國合作對象依策略需求自主選擇，組合隨回合推進持續調整，反映出動態協商的歷程特質。活動後回饋整體正向，「期待」面向平均分數最高，顯示學生對後續類似活動具高度參與意願。 綜合而言，〈烏托邦〉系統採議題導向設計，透過虛實整合互動模式與AI輔助機制，提供兼具情境性、互動性與回顧性的學習經驗，在支持議題理解、行動判斷與互動協商方面具有相當潛力，可作為未來議題導向遊戲式學習與數位教學系統設計之重要參考。;This study aims to design and construct an issue-oriented， phygital digital learning system， Utopia. Centered on complex social issues， the system features six virtual nations with distinct resource conditions and developmental characteristics. System functions include real-time display of world and national parameters， declaration publishing， trade negotiation， action card management， and an AI game assistant (Utopia MGM) that evaluates each nation′s decisions and triggers unexpected events. Activities are implemented in a phygital format， where learners negotiate through face-to-face discussion while completing action selections and progress records via the digital system. The study involved 27 fourth-grade students from an elementary school in northern Taiwan， organized into groups of 4 to 5， each representing one of six nations: the Resource Nation， Market Nation， Energy Nation， Consumer Nation， Agricultural Nation， and Development Nation. Over two rounds， participants engaged in event discussions， declaration publishing， trade negotiation， and action card management. The study primarily employed qualitative methods supplemented by quantitative questionnaires， analyzing declaration content and transaction records through a behavioral coding scheme， determining inter-national interaction patterns using a relationship definition coding scheme， and presenting satisfaction questionnaire results through descriptive statistics. Results indicate that learners predominantly adopted a multiple-objective action mode when responding to events， simultaneously pursuing event resolution， parameter management， and trade negotiation. When facing unexpected events， however， learners shifted to a single objective of event resolution， suggesting that global-scale scenarios prompted learners to prioritize problem-solving. Regarding interaction patterns， nations selected cooperative partners autonomously based on strategic needs， with combinations continuously adjusted across rounds， reflecting a dynamic negotiation process. Post-activity feedback was broadly positive， with the "expectation" dimension receiving the highest average score， indicating strong willingness to participate in similar activities in the future. In summary， the Utopia system adopts an issue-oriented design， delivering contextual， interactive， and reflective learning experiences through a phygital interaction mode and AI-assisted mechanisms. The findings demonstrate considerable potential in supporting issue comprehension， action judgment， and interactive negotiation， offering a valuable reference for future designs of issue-oriented game-based learning and digital instructional systems.
