多模態的具身情境學習及反思回顧系統;Multimodal Embodied Situated Learning and Reflection System

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Computer Science and Information Engineering > Electronic Thesis & Dissertation > Item 987654321/95626

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/95626

Title:	多模態的具身情境學習及反思回顧系統;Multimodal Embodied Situated Learning and Reflection System
Authors:	任家駿;Ren, Jia-Jyun
Contributors:	資訊工程學系
Keywords:	情境認知;數位自我;具身認知;空間運算;多模態;鷹架理論;反思學習;Digital Self;Embodied Cognition;Multimodality;Reflective Learning;Scaffolding;Situated Cognition;Spatial Computing
Date:	2024-07-26
Issue Date:	2024-10-09 17:06:35 (UTC+8)
Publisher:	國立中央大學
Abstract:	傳統的學習方法大多依賴於紙本教材，並由老師口頭教導，形成了老師主導教學，學生在教室內被動接受知識。然而在技能密集型的學科如體育運動、餐旅服務等實作學科中，學習者需要透過自身的多種感官及肢體行為進行知識的學習與實際應用，並於學習後對其多模態的行為進行反思。為了能使學習者能在真實的情境中，透過自身多種具身行為進行學習及反思，本研究設計了一款多模態的具身情境學習及反思回顧系統，並將其應用於數位同步情境中，透過立體視覺相機及麥克風將學習者各種不同的感官表現擷取並儲存下來，使學習者能透過自身多模態的具身行為作為互動基礎，控制虛擬化身進入虛擬情境中進行學習。學習過程中，系統能透過多模態資料對學習者的具身行為進行即時評量並給予相對應的多模態學習鷹架及系統回饋，且學習者能透過觀看多模態的學習紀錄進行反思學習，使學習者在不提升認知負荷的情況下，提升其學習成效、學習動機、反身性思考。為了驗證此系統及教學方法是否能提升學習相關面向，本研究進一步將系統應用於實際課程中。研究結果顯示，本研究之系統在不提升認知負荷的情況下，可以提升學習者的學習成效，且學習動機及反身性思考方面也有正向的影響。 ;Traditional learning methods largely rely on printed materials, with teachers delivering knowledge orally, leading to teacher-dominated instruction and passive knowledge reception by students in the classroom. However, in skill-intensive disciplines such as physical education and hospitality services, learners need to engage multiple senses and bodily actions to learn and apply knowledge practically. Furthermore, learners are required to reflect on their multimodal behaviors after learning. To enable learners to engage in learning and reflection through their own embodied behaviors in scenarios, this study has designed a multimodal embodied situated learning and reflection system. Applied in a digital synchronous setting, the system captures and stores the learners′ various sensory performances using stereo cameras and microphones. This allows learners to interact based on their multimodal embodied behaviors and control a avatar to enter virtual scenarios for learning. During the learning process, multimodal data analysis allows for real-time assessment of learners′ embodied behaviors, providing corresponding multimodal learning scaffolds and system feedback. Learners can engage in reflective learning by reviewing their multimodal learning records. With this system, learners can enhance their learning outcomes, motivation, and reflective thinking without increasing cognitive load. To validate whether this system can enhance related learning aspects, the system was further applied in actual courses. The research results indicate that this system can enhance learners′ learning outcomes without increasing cognitive load, and also positively affects their learning motivation and reflective thinking.
Appears in Collections:	[Graduate Institute of Computer Science and Information Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	18	View/Open

社群 sharing

Loading...