運用眼球追蹤技術分析AR環境下的地科概念學習歷程及科學推理表現

Abstract

本研究利用文字、圖片以及擴增實境互動App來介紹熱點火山主題，並透過眼球追蹤技術紀錄受試者的學習歷程，進而分析受試者在閱讀不同表徵時的視覺注意力分配，以及透過閱讀資料進行不同形式的推理表現如何受視覺注意力影響，與科學推理表現和受試者本身的先備知識、自我效能以及科技態度的關係。本研究招募非理工相關科系大二以上的大學生共34位，先填答13題先備知識測驗，再以頭戴式眼動儀了解受試者在閱讀熱點火山學習系統時的視覺注意力模式，並搭配不同形式的科學推理問題，最後請受試者填寫包含自我效能、科技態度向度問卷。綜整所有實驗數據後，透過描述性統計、獨立樣本T檢定、Pearson相關性檢定以及多元線性迴歸分析來得到眼動指標、科學推理表現與態度量表各向度之間的關係，並進一步了解是否可透過指標及學習態度來預測科學推理表現。 研究結果發現先備知識與部分形式的科學推理表現有關，但當受試者對學習材料的視覺注意力越高，文圖訊息整合的越好，科學推理表現受先備知識影響的程度就越低，此情形在解釋、舉證以及自我反思三種形式的科學推理皆有出現。比較不同閱讀媒介中的眼動指標數據可知受試者整體對擴增實境物件給予較多的視覺注意力，也較不需要整合擴增實境與相關主題的文圖訊息，但低先備知識的受試者對擴增實境物件的注意力較高先備知識受試者更多，且其在特定主題仍有整合擴增實境與文圖訊息的需求。此外，受試者的地科學習自我效能及科技自信自評分數越高，給予文本及擴增實境物件的視覺注意力越低；但擴增實境學習態度越高，在擴增實境媒介的視覺注意力越高。整體而言，本研究結果突顯了視覺注意力對科學推理表現的重要性。This study used texts, pictures, and an Augmented Reality interactive Apps to introduce concepts about Hotspot (a kind of volcanic), and used eye tracking technology to record the learning process in the AR system. Learners’ visual attention distributions to texts, pictures and AR were then analyzed using statistical methods. This study also explored the relationship between performance of scientific reasoning and visual attention, and prior knowledge, self-efficacy and conceptions of using technological to learning. The results of the study found that prior knowledge was related to the performance of scientific reasoning, but the effect would be reduced when the participants paid more attention to the experiment information. The results also showed that while most of the participants, paid more visual attention to and have lower inter-scanning count in the AR environment, lower prior knowledge participants, had higher need to integrate the information between AR, texts and pictures. In addition, it was found that the participants with higher self-efficiency of learning earth science and higher confidence of using technology paid less visual attention to reading information, but those who with higher tendency to using AR for learning had more attention to AR media. Overall, the results of this study highlight relationship between visual attention to the performance of scientific reasoning.