學習壓力評估模式建立與驗證之研究

Establishment and validation of learning stress assessment model

蔣宜成(Yi-Cheng Chiang)

指導教授 ： 姜琇森

國立臺中科技大學/資訊與流通學院/資訊管理系碩士班/碩士(2017年)

摘要

許多研究指出壓力是影響學習成果的重要因素。適當的壓力可以激發潛能，提高能力的發揮並增加效率，相反的當壓力過高或過低會使人的心理或生理產生負面影響，進而導致學生學習的效率不佳。然而，每個人能承受的壓力程度將因個體差異而有所不同。因此，如何評估壓力程度來找出不同個體的壓力承受度是一個重要問題。
本研究採用非對稱最小平方法(Asymmetric least squares)、巴特沃斯濾波器(Butterworth filter)與模糊C均值分群法(fuzzy C-means clustering, FCM)等技術，發展學習壓力評估模式。此外，本研究也提出一壓力量化的演算法，利用模糊C均值分群法資料集中的特性，透過最小-最大縮放法(Min-Max scaling)將壓力數值的特徵重新縮放(feature re-scaling)來進行數據量化，以得到不同腦波特徵下的壓力程度量化數值。最後，本實驗設計兩個情境去探討學生在學習壓力下的學習成效：(1)以不同教學方式形成不同程度的學習壓力情境(放任式、回家作業、課後小考)，探討學生的學習成效。(2)設計4個難度不同的學習單元，且每個學習單元難度持續遞增(壓力遞增)的實驗情境，測試受測者的學習成效是否能持續維持，以了解當學習者經學習且能力提升後，抗壓性是否也能夠提高。
實驗發現在不同教學壓力情境下對學習成效有顯著之影響(P<.001)，在回家作業組與隨堂測驗組中壓力將會提高，隨堂測驗組的學習壓力雖低於回家作業組，但學習成效高於回家作業組，且學生在積極正向的學習壓力中有較佳的學習成效。在不同難度學習單元的情境中，自主式與教學式學習過程會產生不同的學習壓力狀況，在教學式學習過程中，學生對教師的教學依賴較高，會導致面對困難任務時產生較高的學習壓力，而自主式學習的學生在程式語言能力提高後，學生再面對更高難度的任務時，能夠提高自我壓力負荷的程度。此外，當學生面對自我能力能夠負荷的難度時，抗壓性將會提升，而學生有反應難度過高超出能力負荷的難度時壓力將會提升。透過壓力的數值量化可以協助老師與學生本身，有效的管理壓力並找出最佳化的學習效率方式。

關鍵字

壓力 ； 心智負荷 ； 學習成效 ； 腦電圖 ； 模糊C均值分群法

並列摘要

Research suggests that stress is an important factor affecting the learning outcomes. Appropriate stress can stimulate a person to higher performance, increasing your cognitive abilities, and improved working efficiency, but too much stress will have a negative impact on people's physical or mental aspects, resulting in poor learning outcomes. However, the effects and tolerance of stress differs from person to person. Therefore, how to assess the degree of stress to find different individuals’ stress tolerance is an important issue.
This study develops an assessment model to detect the level of stress for learners by using asymmetric least squares, Butterworth filter and fuzzy C-means clustering (FCM). Moreover, this study also proposes a stress quantization algorithm to evaluate the levels of stress on learners. The quantization of levels of stress is carried out for learners under different brain waves features conditions by using the fuzzy C-means clustering and Min-Max scaling. Two experimental scenarios will be designed: (1) different teaching stress scenarios (laissez-faire teaching, homework and quizzes) × learning outcomes. (2) We will design four learning units, and the content of each learning unit will increase the difficulty in order to detect whether the learner's stress tolerance improves with learning process.
The study found that there was a significant effect on learning outcomes under different teaching stress scenarios (P<.001). The level of learning stress in homework group and the quizzed group are higher than the laissez-faire teaching group. Compare with other two groups, the quizzed group have a higher willingness to learn programing course and a good learning achievement. In the case of different difficulty learning units, self-learning group have higher levels of stress in the early learning than auxiliary learning group. The research finds that with increasing the difficulty of each learning unit, the stress tolerance of self-learning students improves with learning ability. However, the auxiliary learning students rely heavily on the teaching assistance and feel hard to facing difficult tasks. In addition, when learners face the difficulty of learning content is that they can afford, the stress tolerance of learners will be enhanced with their ability in programming language. Finally, the research results will help learners manage stress effectively and find the best way to learn programming language.

並列關鍵字

Stress ； Mental load ； Learning effectiveness ； EEG ； Fuzzy C-means clustering

參考文獻

Anyanwu, G. E., Nto, J. N., Agu, A. U., Ekezie, J., & Esom, E. A, “Musical preferences and learning outcome of medical students in cadaver dissection laboratory: A Nigerian survey,” Annals of Anatomy-Anatomischer Anzeiger, Vol. 208, pp. 228-233, 2016.

Chen, S. A., & Desmond, J. E., “Cerebrocerebellar networks during articulatory rehearsal and verbal working memory tasks,” NeuroImage, Vol. 24, No. 2, pp. 332-338, 2005.

Cho, Y., Kim, J., & Yang, M. “Investigation On Structural Relationship Among Adolescents’ Media Use, Self-Directed Learning, And Ego-Resilience,” Procedia-Social and Behavioral Sciences, Vol. 191, pp. 1426-1430, 2015.

Choi, Y., Kim, M., & Chun, C. “Measurement of occupants' stress based on electroencephalograms (EEG) in twelve combined environments,” Building and Environment, Vol. 88, pp. 65-72, 2015.

Davidson, R. J. “11 Affect, cognition, and hemispheric specialization,” Emotions, cognition, and behavior, pp. 320, 1988.