| 摘要: | 理解學生的注意力，特別是視覺注意力，對於有效的教學與學習至關重要，因為它充當過濾機制，決定哪些資訊會被處理。社會臨場感(Social presence) 同樣在塑造學生對課程內容的投入方面發揮重要作用。儘管許多研究已經探討了教師在螢幕上的存在感，但對於其在實體教室中的影響則知之甚少，主要原因是缺乏如可穿戴式眼動追蹤系統這類準確、即時的監測工具。此外，在多媒體環境中，教師的手勢與動作如何引導學生的注意力由文本轉移到圖像，在實體教室中仍然不清楚。另一方面，人們對於在自主程序性學習的過程中理解視覺注意力的興趣日益增加，特別是整合智慧回饋與觸控互動(例如觸控圖示)時。
為了填補這些研究缺口，我們遵循PRISMA(Preferred Reporting Items for Systematic Review and Meta-Analysis, PRISMA)指南進行了系統化文獻回顧。本次回顧探討了與教師存在感及非語言行為相關的研究趨勢、主題發展和關鍵字模式，特別關注使用眼動追蹤方法的研究。研究結果顯示，儘管教師的存在感與非語言提示可以對學生的視覺注意力產生顯著影響，但它們對於學習成果的影響程度會因教育環境以及教學設計而異。此外，本次回顧還顯示了各項研究之間的重要主題關聯性，強調了教師存在感與視覺注意力在學習過程中的關鍵作用。本領域的主要研究趨勢包括注視、視覺注意力、線上學習、影片講解及社會臨場感，這些議題近年來成為研究的焦點。此外，本次回顧還辨識出幾個重要的研究群體，主要圍繞在多媒體學習原則、認知負荷理論以及教師行為的影響等議題。
基於這些研究發現，第一項研究專注於實體教室環境中的教師存在感。本研究採用準實驗設計，對25名研究生進行實驗，並利用可穿戴式眼動追蹤系統比較兩種教學模式：教師現實體授課與錄製影片線上授課。透過熱圖分析注視點分佈，顯示受試者傾向花費更多時間將注意力集中在以文字為主的簡報上。值得注意的是，教師的手勢與身體動作在引導學生的視線離開文字簡報方面特別有效，突顯了非語言提示在引導注視點中的關鍵作用。此外，現場教師的存在使學生更專注於教師的臉部，而非身體的其他部位，進一步說明社交提示在實體授課中的強大影響力。
第二項研究探討了基於模擬的COVID-19快篩系統中的自主程序性學習，該系統整合了智慧回饋、眼動追蹤與觸控技術。透過同步分析眼動數據與觸控互動，我們研究了受試者的手眼協調與視覺注意力。60名研究生參與了互動式的程序性任務，並使用滯後序列分析(Lag Sequence Analysis, LSA)來捕捉眼動數據與任務相關行為之間的相互作用。研究結果顯示，獲得即時智慧回饋的受試者錯誤率較低，能更有效率地進行操作，整體表現顯著優於未獲得回饋的受試者。此外，智慧回饋顯著提升了學習成效，使學生獲得更高的成就，並且比只獲得一般回饋的受試者更有效率地完成任務。
整體來說，這些研究結果強調了教師存在感與非語言溝通在吸引學生視覺注意力方面的重要性，以及智慧回饋在提升學習成效與任務效率方面的有效性。本研究開發的可穿戴式眼動追蹤系統為探討學生在實體教室環境中的視覺注意模式提供了新的可能性，而熱圖等視覺化技術在呈現眼動數據方面也發揮了關鍵作用。此外，滯後序列分析提供了一種穩健的方法來理解學生如何引導他們的注視點並與觸控式任務進行互動。總之，這些研究發現凸顯了在教學實踐中整合教師提示與智慧回饋機制的必要性，以提升學生的學習參與、加深理解，並促進自主學習。
;Understanding student attention, particularly visual attention, is crucial for effective teaching and learning, as it functions as a filtering mechanism that determines which information is processed. Social presence also plays a significant role in shaping students’ engagement with course content. Although many studies have examined instructor presence on-screen, far less is known about its impact in physical classrooms, largely due to the lack of accurate, real-time monitoring tools like wearable eye-tracking systems. Additionally, how instructors’ gestures and movements guide learners’ attention from text to images in multimedia settings remains unclear in the physical classroom. There is also growing interest in understanding visual attention during self-guided procedural learning, especially when smart feedback and touch-based interactions (e.g., touch plots) are involved.
To address these gaps, we conducted a systematic literature review following PRISMA guidelines. This review explored research trends, thematic developments, and keyword patterns related to instructor presence and nonverbal behaviors in relation to student attention, particularly in studies using eye-tracking methods. Findings indicate that while instructor presence and nonverbal cues can meaningfully influence learners’ visual attention, the extent of their impact on learning outcomes varies according to educational contexts and instructional design. The review also reveals significant thematic connections among studies, emphasizing the critical role of instructor presence and visual attention. Key trends in this field include gaze, visual attention, online learning, video lectures, and social presence, all of which have recently become focal points of research. Additionally, the review identifies prominent research clusters centered on multimedia learning principles, cognitive load theory, and the influence of instructor behaviors.
Building on these findings, the first study focused on instructor presence in a physical classroom environment. A quasi-experimental design was conducted with 25 graduate students, employing a wearable eye-tracking system to compare two instructional modes: live instruction by an instructor and video-recorded instruction. Heat map analyses of fixation distributions revealed that, during video-recorded instruction without an instructor, participants spent more time focusing on text-based slides. Notably, the instructor’s gestures and body movements were particularly effective in redirecting learners’ attention away from text slides, underscoring the critical role of nonverbal cues in guiding visual focus. Moreover, the presence of a live instructor prompted students to focus more on the instructor’s face than on other parts of the body, highlighting the powerful influence of social cues in live instruction.
The second study examined self-directed procedural learning in a simulation-based COVID-19 rapid test system that integrated smart feedback and touch-based eye-tracking. By synchronizing eye movement data with touch interactions, we analyzed learners’ hand-eye coordination and visual attention. Sixty graduate students engaged in interactive procedural tasks, and Lag Sequence Analysis (LSA) were employed to capture the interplay between eye-tracking data and task-related actions. Results showed that participants who received real-time smart feedback made fewer errors and completed manipulations more effectively, demonstrating higher overall performance than those without such feedback. Moreover, smart feedback substantially improved learning outcomes, leading to greater achievement and more efficient task completion compared to basic feedback.
Overall, these findings underscore the importance of instructor presence and nonverbal communication in capturing students’ visual attention, as well as the effectiveness of smart feedback in enhancing learning outcomes and task efficiency. The wearable eye-tracking system developed in this research offers novel opportunities to investigate learners’ visual attention patterns in authentic classroom settings, with visualizations such as heatmaps proving useful for representing eye-tracking data. Moreover, LSA provides a robust method for understanding how learners direct their gaze and interact with touch-based tasks. Collectively, these outcomes highlight the need to integrate instructor cues and smart feedback mechanisms into pedagogical practices to bolster student engagement, deepen comprehension, and foster self-directed learning. |
