| 摘要: | 越來越多非英語系國家的大學推行將英文門檻納入學校的畢業門檻,不僅在中小學以及高等教育中更加著重英語教學,甚至出國念書與交換也日漸普及,為了提高出國念書與交換的機會,聽說讀寫能力缺一不可,因為聽、說、讀、寫是雅思(IELTS)和托福(TOEFL)這兩種英語檢定測試的四大方向,之前歐洲國家學校多採認雅思;美國學校多採認托福,現今歐美國家大部分學校都採認雅思成績為出國留學與交換的英語門檻,雅思對準備出國留學學生的重要性日漸升高,但一味學習的過程可能會感到枯燥乏味,缺乏互動性從而造成學生缺乏學習動機,甚至導致學習效果日漸下滑,因此不少研究將英語學習及數位遊戲式學習(DGBL)結合設計成系統,以提升學習成效及學習動機。在數位遊戲式學習環境中,引入故事情境、任務挑戰、學習策略與排
行榜等遊戲元素,以提升學習趣味性與參與度,激發學習者的學習動機,讓學習者有持續使用系統的動力,然而近期關於數位遊戲式學習的研究結果顯示,數位遊戲式學習仍存在個別化不足與學習歷程追蹤困難等問題,過去研究顯示,每位學習者皆具有個體差異,個性化回饋可能成為學習歷程中重要的成敗因素,過去研究顯示,人工智慧(AI)技術能解決學習者無法獲得即時、精準的回饋與引導等問題,生成式人工智慧(GenAI)模型,能利用過去積累的大量資料,以引導式的訓練生成方式,將其運用於問題的解析,達成強化學習行為分析、內容推薦與動態回饋功能等目標,從而提升學習適應性與成效。
本研究設計不同學習系統版本,包含傳統模擬、傳統模擬模式與 DGBL 結合、AI與 DGBL 結合模式,分別為傳統解析模式模擬系統、數位遊戲式學習的傳統解析模式、數位遊戲式學習的智能解析模式。研究對象共 77 位學習者,平均分配至三組不同學習策略的系統版本。本研究旨在不同學習策略的雅思閱讀學習環境中,探討不同學習策略對雅思閱讀學習的學習成效、遊戲成效、學習動機、學習觀感之影響及相關性,比較三種系統對高中生英語學習成效與學習動機之影響。結果顯示,AI 與 DGBL 整合系統在學習表現與動機提升方面皆顯著優於其他兩組,顯示智慧化遊戲式學習具高度教育潛力與發展價值。
本研究結果顯示,在不同學習策略的雅思閱讀學習環境下對學習者的影響。本研究成果分為以下三點:(1) 數位遊戲式學習的智能解析模式不僅雅思閱讀學習成效有良好的表現,在學習動機及學習觀感上也表現最好。(2)數位遊戲式學習的傳統解析模式有較好的英語學習成效及遊戲成效外,在學習動機及學習觀感上也表現較好。 (3)相關性的部分,三組不同學習策略的雅思閱讀學習成效與遊戲成效之間皆有較多的顯著相關,三組的學習動機與學習觀感皆有顯著相關,本研究更進一步的結果,發現三組不同學習策略下的遊戲成效、學習動機和學習觀感之間的相關性也有所不同。本研究的研究成果提供不同學習策略對學習者其影響,這些結果有助於未來研究及教學使用,提供未來的研究者一個以不同學習策略為主的學習系統的設計架構。;More and more universities in non-English-speaking countries are incorporating English as a graduation threshold. Not only is English teaching emphasized in primary and secondary
schools as well as in higher education, but it is also becoming more and more popular to study and exchange abroad. In order to increase the chances of studying and exchanging abroad, the ability to speak, read and write is indispensable because listening, speaking, reading and writing are the four major directions of the English language tests of the International English Language
Testing System (IELTS) and the Test of English as a Foreign Language (TOEFL). The importance of IELTS to students preparing to study abroad is increasing, but the learning
process may be boring and lack of interactivity, resulting in a lack of motivation for students to learn, and even leading to a gradual decline in the learning effect. Therefore, many studies have been conducted to design systems that combine English language learning and digital game-based learning (DGBL) to enhance learning performance and motivation. In a DGBL environment, game elements such as story situations, task challenges, reward and punishment
mechanisms, and leaderboards are introduced to enhance the interest and participation in learning, stimulate learners′ motivation, and motivate learners to continue using the system. However, recent studies on DGBL have shown that there are still problems with the lack of individualization and difficulties in tracking the learning process. Past research has shown that each learner has individual differences, and personalized feedback can become an important success factor in the learning process. Past research has shown that artificial intelligence (AI)
technology can solve the problem of learners not being able to get immediate and accurate feedback and guidance, and generative artificial intelligence (GenAI) models can utilize the past to provide a more accurate feedback and guidance. GenAI models can make use of a large amount of data accumulated in the past to generate guided training and apply it to problem solving to achieve the goals of strengthening learning behavior analysis, content
recommendation, and dynamic feedback functions, thereby enhancing learning adaptability and effectiveness.
In this study, different versions of learning system are designed, including traditionalsimulation, traditional simulation mode combined with DGBL, and AI combined with DGBL, which are the Traditional Analytical Simulation Mode System (TASMS), the Digital Game-Based Learning with Traditional Analytical Mode System (DGBL-TAMS), and the Digital Game-Based Learning with Intelligent Analytical Mode System (DGBL-IAMS). A total of 77 learners were equally assigned to the three versions of the system with different learning system modes. The purpose of this study was to investigate the effects and correlations of different
learning system modes on learning effectiveness, In-game performance, learning motivation, and learning perception of IELTS reading learning in an IELTS reading learning environment with different learning system modes, and to compare the effects of the three systems on the learning performance and learning motivation of senior high school students in English language learning. The results show that the AI and DGBL integration systems significantly
outperform the other two groups in terms of learning performance and motivation, indicating that intelligent game-based learning has high educational potential and developmental value.
The results of this study show the effects of different learning system modes on learners in the IELTS reading learning environment. The results of this study are categorized into the following three Score: (1) The intelligent parsing model of digital game-based learning not only has good performance in IELTS reading learning outcomes, but also has the best performance in learning dynamics and learning perception. (2) The traditional parsing model of digital game-based learning not only has better English learning outcomes and game outcomes, but also performs better in learning dynamics and learning perception. (3) In the correlation part, there were more significant correlations between the IELTS reading learning outcomes and game outcomes in the three groups with different learning system modes, and there were significant
correlations between learning motivation and learning perception in the three groups. Further results of this study found that the correlations between game outcomes, learning motivation and learning perception in the three groups with different learning system modes were also
different. The results of this study provide information about the effects of different learning system modes on learners, which will be useful for future research and teaching, and provide future researchers with a framework for designing learning systems based on different learning
system modes. |
