- 期刊
- OpenAccess
五年級資優生與專家使用圖形化程式（NXT-G）之心智模式及建模歷程
Fifth Grade Gifted Students' and Experts' Mental Models and Modeling Processes in NXT-G
摘要
本研究意欲比較五年級資優生與專家對行動機器人Lego Mindstorms圖形化程式（NXT-G）中「迴圈」與「分岔」兩程式進行概念理解與設計時，可能產生的心智模式與建模歷程。資優生可視為學習之起點，專家則為可能的學習典範，藉此比較，希望對資優生行動機器人程式之教學提出建議。研究以個案研究方式選擇六位初學NXT-G之五年級資優生及兩位專家，透過螢幕攝影及放聲思考，收集其進行「程式概念檢測題」之過程，並輔以「程式概念理解測驗」與晤談之結果進行歸納與推論。結果發現：一、在心智模式方面，資優生於迴圈程式中具有「loop」及「step」模式，而分岔程式則為「singleswitch」、「multi-switch」及「odd」模式；專家在迴圈程式中呈現「variable loop」及「nested loops」模式，而分岔程式則為「multi-switch」模式。二、在建模歷程方面，資優生的建模歷程通常依賴上課經驗選擇模型，且選定後，需執行實驗方能驗證模型，若未能達到目的，便多以試誤方式微調參數或拒絕模型而無法修正；專家則具備豐富解題經驗與資源，形成策略模組，並多運用思考實驗驗證所選之模型，解題快速。研究者建議運用variable loop、nested loops等專家策略模組形成版型加強學生基本概念及使用程式模組經驗，並多運用「預測－執行－分享」鼓勵學生思考實驗以比較各種模型的範圍與限制。
並列摘要
This study explores fifth grade gifted students' and experts' mental models and modeling process, in the development of conceptual understanding and design of ＂loop＂ and ＂switch＂ programs in NXT-G. The students' performance could be seen as a starting point, and experts' performance could be seen as the role model. Suggestions for gifted students learning mobile robotics were proposed by comparing these two target groups. This study adopted the case study research method. Six fifth grade gifted students and two experts were purposely selected. Participants' experiences in completing the ＂Programming Concepts Test＂ were collected through the think aloud method and screen video capture. Additional evidence was gathered from the ＂Test of Understanding in Programming Concepts＂ and interviews. The results showed: 1. The ＂loop＂ and ＂step＂ models were adopted by the gifted students when they understood the ＂loop＂ program, while ＂single-switch,＂ ＂multi-switch＂ and ＂odd＂ models were adopted in students' understanding of the ＂switch＂ program; however, ＂variable loop＂ and ＂nested loops＂ models were adopted by the experts when they performed the ＂loop＂ program, while the ＂multi-switch＂ model was adopted in experts' performance of the ＂switch＂ program. 2. The gifted students relied on class experience to select a model, and needed to verify their model via implementing the program. If they could not achieve the goal, they usually used trial and error to modify arguments or rejected their original model rather than revising it. However, the experts had abundant problem solving resources and experiences to form strategies. They usually verified their selected model via thought experiments and solved problems rapidly. We suggest adopting experts' strategy models (e.g., variable loop, nested loops) as templates to enhance students' basic understanding of programming and their experience of using programming models. Moreover, ＂prediction-implementation-sharing＂ could be used to encourage students' thinking experimentations and to compare different models' scopes and limitations.