【關鍵字】：海洋議題素養、生成式AI、AI素養、知識概念圖;The United Nations proposed the 2030 Sustainable Development Goals， among which Goal 14， "Life Below Water，" emphasizes the conservation and sustainable use of oceans and marine resources. However， marine socio scientific issues possess a high degree of complexity. When facing a vast amount of online information， upper-elementary school learners often struggle to filter it effectively， resulting in their critical thinking and argument construction remaining at a superficial level. To address this dilemma， this study introduced generative AI as a supporting tool to guide learners through issue based inquiry and information analysis， exploring the impact of this instructional intervention on ocean literacy， AI literacy， and concept map construction. This study aims to answer the following questions: 1. What is the effectiveness of the curriculum in enhancing learners′ ocean literacy? 2. What is the effectiveness of the curriculum in enhancing learners′ AI literacy? 3. What is the impact of AI tools on learners′ construction of marine issue concept maps?
This study engaged upper-elementary school learners from an elementary school in northern Taiwan as research participants. A systematic curriculum， titled "Smart Mapping the Ocean，" was designed and implemented. The instructional content included scenario simulations in prerequisite courses， operational training on AI tools and digital concept mapping platforms， and deep critical inquiry into socio-scientific issues within the core curriculum. The core instructional design guided learners to utilize AI tools through a two-round inquiry model， assisting them in filtering evidence and analyzing multi perspective stances amidst complex information， subsequently transforming their discussion results into visual concept maps. In addition， this study employed an ocean literacy test， an AI literacy scale， and a concept map rubric as research instruments to evaluate learners′ quantitative scores and the evolutionary effectiveness of their knowledge structures.
The research results indicate the following: First， the proposed instructional model significantly enhanced learners′ ocean literacy， particularly with performance stabilizing in the dimensions of marine pollution and resources， confirming that the instructional intervention yielded substantial effects on deepening learners′ core marine cognition. Second， learners demonstrated significant progress in AI literacy， with the growth in "Human Centered Mindset" and "AI System Design" dimensions being the most prominent， indicating that the instructional intervention effectively cultivated learners′ technology application and systems thinking capabilities. Third， through the real-time support of AI tools， learners demonstrated a higher degree of structured logic in constructing concept maps， successfully transforming fragmented information into knowledge structures with coherent logical connections. This study confirms that through a systematic instructional model， AI tools can effectively serve as a cognitive scaffolding， assisting learners in overcoming comprehension barriers related to complex issues， and providing concrete instructional guidelines and reference foundations for marine education practices under digital transformation.
Keywords：Ocean Literacy， Generative AI， AI Literacy， Concept Map
