以眼球追蹤技術探究互動學習軟體對七年級學生基因概念的學習

國中的基因與遺傳是教師和學生都覺得難教難學的主題。基於現在資訊輔助教學的普及與國內相關中文學習APP的缺乏，本研究發展一套互動學習軟體，結合眼球追蹤技術、學習者特性問卷探討學生使用基因與遺傳互動軟體的學習。研究對象為41名自願參加的國一學生，於109年4月至7月實驗。前測後，學生進行互動軟體學習，同時收集眼動資料，接著填寫學習者問卷（認知負荷、生物與技術態度及學習動機量表）。實驗後，依後測平均分組：高學習成就組（專家）23人、低學習成就組（生手）18人。針對前後測分數、眼動資料及學習者特性，以SPSS23進行描述性統計、成對樣本t檢定、獨立樣本t檢定、Partial correlation、Pearson correlation與逐步回歸分析。結果發現：前後測的結果發現互動軟體對於學生的基因與遺傳的學習有幫助，且對於高學習成就組的表現稍高。此外，互動軟體的第一部分基礎學習不會造成不同學習成就學生的眼動差異與認知負荷評分的差異，說明互動軟體不會造成不同學習成就學習者過多的認知負荷。此外，從掃視時間看來，高學習成就組需花較花的時間統整第一部分基礎學習的訊息達接近顯著水準，與前人研究相符值得注意。而第二部分進階學習中高學習成就的學習者會降低圖及提示訊息的注意力分配，把注意力多分配在問題理解與答案的思考上。互動軟體的靜態與動態圖來看，互動的動態圖能幫助學生達到較好的學習成就。試題分析來看，高學習成就的學生使用互動軟體進行概念學習後，後測中各個認知歷程－學習內容面向大多都有顯著進步。注意力模式與學習者特性預測學習結果以微觀符號的圖上的掃視時間（SD）預測學習成就的模型配適度較好，此結果說明學習者在圖形上的訊息統整是預測學習成就的主要因素。研究者對未來基因與遺傳的多媒體輔助教學的研究，建議可以就巨觀、微觀、符號層次中選一或二層次概念研究，或是每位受測者的學習提升至20－30分鐘左右。此外，未來也可探討學習者在微觀、巨觀和符號概念或是圖、文、影片等表徵轉換的困難和學習歷程的關係，或比較都市文教區學校與非都市文教區學校的學生學習差異。

關鍵字

基因；遺傳；互動學習軟體；眼球追蹤

並列摘要

In middle schools, teachers and students find it difficult to teach and learn the topics about gene and genetics. Although the technology-assisted teaching is getting popular, there is still a lack of interactive learning apps focusing on the topics of gene and genetics in Taiwan, this study developed an interactive learning software and used eye tracking technology and learner characteristics questionnaires to explore students' learning in this interactive software. The subjects of the study were 41 7th grade students who participated in the experiment from April to July 2020. After the pre-test, students played with the interactive learning software while their eye movements were recorded by the Tobii x3-120 eye tracking system. After the learning activity, participants were asked to fill out the learner questionnaire (assessing Cognitive Load, Biological and Technological Attitudes and Learning Motivation Scale). After the experiment, Students were divided into two groups according to the result of the post-test: 23 people were assigned to in the high learning achievement group (expert) and 18 people in the low learning achievement group (novice). Data analysis methods included descriptive statistics, paired sample t test, independent sample t test, partial correlation, Pearson correlation, and stepwise regression analysis. The result are summarized as followed. A comparison between the pre-test and post-test results suggested that the interactive software was helpful for the learning of students' genes and genetics. No differences in the eye movement patterns and scores cognitive load were found between students with different learning achievements in the first part of the learning program focusing on the basic concepts did not result in cause. The finding indicated that the interactive software did not cause excessive cognitive load on learners. The eye movment analysis showed that the high achievers spent more time to integrate the first part of the basic learning information, which is consistent with previous research. In the second part of learning which involved more advanced concepts, high-achieving learners attended less to the pictures and prompt messages, and allocated more attention to question and answers. Attention patterns and learner characteristics were found to predict learning results. Additionally, it was found that attention to the interactive dynamic diagrams could help students to achieve better learning achievements. The test items were futher differentiated in to different categories based on Bloom’s Taxonomy. By analyzing the test performances, it was found that high achievers performed better in all the content aspects of the post-test. Regression analysis suggested that the saccade duration (SD) on the micro-symbol graph predicted learning. The study result implies that the research on the teaching of gene and genetics may involve one or two levels of concepts divided into macro, micro, and symbolic levels. The learning duration should be kept in 20-30 minutes. In the future, we can also explore the relationships between learners’ difficulties in the conversion of micro, macro, and symbolic concepts or representations of pictures, texts, films, etc. and their learning process, or compare the learning differences between schools in different districts.