優化組織圖對誘人細節效應的作用：生成性學習機制的啟示

Abstract

誘人細節是指教材中與學習目標無關但有趣的資訊。儘管這類資訊可以提升學習者的興趣，但它們也會引發誘人細節效應。為減輕其負面影響，以往研究嘗試應用了多媒體教學設計原則，但效果不佳。本研究以高中生為研究對象，從生成性學習理論的視角出發，探討生成性學習活動之一的組織圖在生物學科說明文中對誘人細節效應的影響。組織圖是一種視覺化生成性學習活動，能將文本轉化為空間視覺形式，通過展示概念、術語或觀點之間的關係促進學習。傳統組織圖分為兩類：向學習者呈現完整的組織圖（filled-in graphic organizers, FGO），以及要求學習者根據文本生成組織圖（interactive graphic organizers, IGO）。儘管FGO直觀且便於展示知識重點，但學習者可能僅僅記憶內容而缺乏深層處理；而IGO則能增加學習者的認知投入，但可能因材料複雜或學習者能力不足，生成的組織圖品質較低。因此，本研究有兩個目標：第一，將對傳統組織圖使用方式進行優化設計，考察優化組織圖是否比傳統組織圖效果更好。第二，考察優化組織圖是否能夠降低誘人細節效應，以及在這個過程中學習者的情境興趣水平會有什麼變化。本研究包含一個前導實驗與兩個正式實驗。前導實驗確定實驗材料的可讀性，篩選難度和區分度適合的前後測題目。實驗一使用信號原則中的多媒體整合信號和分段原則優化傳統形式FGO，幫助學習者更有效的整合組織圖和文本資訊；通過分段提供組織圖框架和關鍵概念，降低傳統IGO生成難度，提升其準確性。211名高中生學習了關於病毒感染人體細胞的說明文，分為四個實驗條件：傳統FGO組、優化FGO組、傳統IGO組和優化IGO組。結果顯示，優化FGO組在保留與遷移測驗上的表現顯著高於傳統FGO組，優化FGO組的學習時間和認知負荷高於傳統FGO組，表明優化FGO促進了更深層次的處理；優化IGO組在保留與遷移測驗上的表現顯著高於傳統IGO組，優化IGO組的學習時間和認知負荷低於傳統IGO組，且組織圖完成的正確率更高，表明優化IGO降低了任務難度，提升了完成品質；優化IGO組在保留和遷移測驗上高於優化FGO，優化IGO組的學習時間和認知負荷高於優化FGO組，證明優化IGO更好地促進了生成性學習過程，提升了學習表現。實驗二中，210名高中生使用與實驗一相同的學習材料，並添加了與病毒相關的誘人細節，分為四個條件：控制組（無組織圖和誘人細節）、誘人細節組、優化FGO加誘人細節組、優化IGO加誘人細節組。首先，結果再次證實了誘人細節效應，即誘人細節組在保留和遷移測驗上低於控制組，外在認知負荷和情境興趣高於控制組，但學習時間無差異。第二，研究發現優化組織圖能降低誘人細節效應，優化FGO加誘人細節組和優化IGO加誘人細節組在保留和遷移測驗中的表現顯著高於誘人細節組。第三，優化FGO加誘人細節組在保留測驗上的表現高於優化IGO加誘人細節組，兩者在遷移測驗上無差異。第四，實驗二還發現外在認知負荷的中介作用，即優化FGO通過降低外在認知負荷提升學習表現。最後，研究發現在誘人細節情境下使用組織圖，學習者的情境興趣會下降。兩個實驗結果表明，相較於傳統方法，優化組織圖能更好地促進生成性學習過程，提升學習表現。學習材料的性質會影響優化FGO和IGO的效果，尤其當材料中包含誘人細節時，優化FGO是更好的選擇。本研究的局限在於使用圖形組織者的形式比較單一，未來研究可以嘗試考察其他類型的組織圖對生成性學習有何種影響，或者可以嘗試將優化FGO與優化IGO組合使用更好的發揮優化組織圖對於學習的促進作用。其次，未來研究也可以考慮增加對於認知過程的直接測量，例如眼動技術，進一步解釋優化組織圖是如何降低誘人細節效應的。最後，本研究僅採用了立即測驗，未來研究可以嘗試通過增加延宕測驗來考察誘人細節潛在的情緒動機作用。Seductive details refer to interesting pieces of information within an expository text that are only tangentially related to the target concept. While such information can improve learners' learning interest, it also raises extraneous cognitive load. When the presence of this information results in reduced comprehension, this is called the seductive details effect. Previous research has attempted to mitigate the negative impact of seductive details by applying multimedia learning principles to reduce the extraneous cognitive load they induce, but with limited success. This study explores the impact of a generative learning activity, specifically graphic organizers, on the seductive details effect from a generative learning perspective.Graphic organizers are a visual generative learning activity that transform text into spatial-visual formats, promoting learning by showing the relationships between concepts, terms, or ideas. Graphic organizers can be categorized into two types: fully completed graphic organizers (FGO) presented to learners, and interactive graphic organizers (IGO) that require learners to generate them based on the text. Although adding a matrix organizer provides a better opportunity for generative learning, the student may still just try to memorize the material in the matrix, without engaging in deeper processing. IGOs, on the other hand, increase cognitive engagement but might result in low-quality graphic organizers due to material complexity or insufficient learner abilities. Therefore, this study has two objectives: Firstly, to optimize the traditional FGOs and IGOs and examine whether the optimized versions perform better than traditional ones; secondly, to investigate which type of optimized graphic organizer can reduce the seductive details effect and the roles of extraneous cognitive load t in this process.This study includes a prior study and two main experiments. The prior study aims to determine the practicality of the experimental materials and select pre- and post-test questions with appropriate difficulty and discrimination. Experiment 1 optimizes traditional FGOs using signaling and segmenting principles to help learners integrate graphic organizers and text information, and reduces the difficulty of generating IGOs by segmenting the framework and key concepts. Two hundred high school students studied an expository text about viral infections of human cells, divided into four conditions: Traditional FGO group, optimized FGO group, traditional IGO group, and optimized IGO group. Results showed that optimized graphic organizers performed better on retention and transfer tests; the optimized FGO group had higher learning time and cognitive load than the traditional IGO group, indicating that optimized FGOs promote deeper processing; the optimized IGO group had lower learning time and cognitive load than the traditional IGO group, with higher accuracy in completing graphic organizers, indicating that optimized IGOs reduced task difficulty and improved quality; the optimized IGO group outperformed the optimized FGO group in retention and transfer tests, suggesting that optimized IGOs better facilitate the generative learning process and enhance learning performance.In Experiment 2, 210 high school students used the same learning materials as in Experiment 1, with seductive details related to viruses added to the seductive details and optimized graphic organizer groups. They were divided into four conditions: Control group (no graphic organizers or seductive details), seductive details group, optimized FGO + seductive details group, and optimized IGO + seductive details group. First, the results confirmed the seductive details effect, with the seductive details group performing worse on retention and transfer tests than the control group, and having higher extraneous cognitive load and situational interest but no difference in learning time. Second, the study found that optimized graphic organizers can reduce the seductive details effect, with both optimized FGO + seductive details and optimized IGO + seductive details groups outperforming the seductive details group in retention and transfer tests. Third, the optimized FGO + seductive details group performed better on retention tests than the optimized IGO + seductive details group, with no difference in transfer tests; the optimized FGO + seductive details group had lower learning time and extraneous cognitive load than the optimized IGO + seductive details group, contrary to Experiment 1. Fourth, experiment 2 also found a mediating effect of extraneous cognitive load, with optimized FGOs enhancing learning performance by reducing extraneous cognitive load. Finally, the study found that using graphic organizers in seductive details contexts leads to a decrease in learners' situational interest. These results indicate that compared to traditional methods, optimized graphic organizers better facilitate cognitive processes and enhance learning performance. The complexity of learning materials affects the effectiveness of optimized FGOs and IGOs, especially when seductive details are included, making optimized FGOs a better choice. The limitations of this study lie in the relatively singular form of graphic organizers used. Future research could explore the effectsof other types of graphic organizers on generative learning or attempt to combine optimized FGO and optimized IGO to better enhance learning. Additionally, future studies could consider increasing direct measurements of cognitive processes, such as using eye-tracking technology, to further explain how optimized graphic organizers reduce the seductive details effect. Lastly, this study only utilized immediate tests; future research could attempt to assess the potential emotional and motivational effects of seductive details by delayed tests.