畫記策略在大學生閱讀說明文本時的作用：眼動追蹤研究

Abstract

畫記（highlighting）作為一種簡便且高頻的閱讀策略，廣泛被學生用於強調資訊重點，然而其對閱讀理解的成效卻仍存有爭議。過去研究多指出畫記對記憶有輔助作用，卻無法顯著促進文本理解的深層歷程，甚至在某些情境下產生負向影響。然而相關研究多使用事後測驗，對於畫記實際閱讀歷程的影響仍缺乏直接觀察。有鑑於此，本研究結合眼動追蹤技術，探討大學生在閱讀長篇說明性文本時使用畫記策略的歷程與其對閱讀理解表現的影響。研究共招募60位大學生（含研究生），隨機分派至畫重點組與純閱讀組，兩組皆閱讀一篇約3600字的科普說明文本。畫重點組需使用電腦滑鼠進行畫重點操作，純閱讀組則僅進行閱讀，無任何標記。閱讀後所有參與者在無法回訪文章的情況下完成文本理解測驗，測驗題目涵蓋文本基礎層次（如直接提取、推論）之理解歷程，並進行閱讀能力與畫記信念的測量。全程以EyeLink Portable Duo蒐集眼動資料，並分析五個指標：總凝視時間、首次凝視時間、重讀時間、重讀次數與佔總凝視時間比率。結果發現，兩組在閱讀理解測驗成績上並無顯著差異，顯示畫記策略在本研究設計下無法直接提升文本細節的理解與回憶，閱讀能力仍是主要影響表現的變項。在調節效果部分，畫記信念與策略使用之間存在顯著交互作用：畫記信念高的參與者若無法使用該策略，則在測驗表現上會顯著下降，顯示策略選擇與個人習慣間的衝突可能帶來認知負擔。眼動分析顯示，畫重點組具有更長的總凝視時間與首次凝視時間，顯示其需投入額外注意資源以執行策略。此外，畫重點組於重點密度高的段落顯示較明顯的注意力集中現象，其凝視時間比例顯著高於純閱讀組，呈現出選擇性加工的視覺歷程；而純閱讀組的注意力分配則較為平均，未出現段落差異。綜合而言，本研究結果支持畫記作為「注意力導引」的角色，但對文本內容組織的幫助可能有限。Highlighting is a simple and frequently used reading strategy employed by students to emphasize key information. However, its effectiveness in enhancing reading comprehension remains controversial. While previous studies have shown that highlighting may support memory, it has not been found to significantly promote deeper-level text comprehension processes, and in some cases, may even lead to negative effects. Notably, most existing research relies on post-reading tests and lacks direct observation of the reading process itself. To address this gap, the present study integrated eye-tracking technology to investigate the reading processes and comprehension outcomes of college students using highlighting strategies while reading a long expository text. Sixty college and graduate students were recruited and randomly assigned to either a highlighting group or a reading-only group. Both groups read a 3,600-word popular science article. The highlighting group used a computer mouse to mark important text, whereas the reading-only group read without any annotations. After reading, all participants completed a reading comprehension test without access to the text. The test items targeted textbase-level comprehension, such as direct retrieval and inferences. Measures of reading ability and beliefs about highlighting were also collected. Eye movements were recorded throughout the reading process using the EyeLink Portable Duo, and five indicators were analyzed: Total fixation duration, first-pass fixation duration, rereading, regression count, and percentage of fixation duration. The results revealed no significant difference in comprehension scores between the two groups, suggesting that highlighting, under the conditions of this study, did not enhance comprehension and recall of textual details. Reading ability remained the primary factor predicting comprehension performance. Regarding moderation effects, a significant interaction was found between beliefs about highlighting and the use of the strategy: Participants with strong beliefs about highlighting performed worse when they were not allowed to use it, suggesting that a mismatch between preferred and assigned strategies may impose a cognitive load. Eye-tracking data showed that the highlighting group exhibited longer first-pass fixation duration and total fixation duration during initial reading, indicating that they needed to invest additional attention resources to implement the strategy. Moreover, this group demonstrated more focused attention in high-density segments of highlighted content, reflected in a significantly higher proportion of fixation duration compared to the reading-only group, revealing a selective visual processing pattern. In contrast, the reading-only group distributed their attention more evenly across text segments. In sum, the findings of this study support the role of highlighting as a tool for directing attention but suggest that its benefits for organizing textual content may be limited.