The Principles of Deep Learning for Cultivating Competencies
——From the Perspective of Coexistence of Knowledge and Methods
(1. Beijing Normal University, China Basic Education Quality Monitoring Collaborative Innovation Center, Beijing 100875;
2. Southwest University, Faculty of Education, Chongqing 400715)
The following video is sourced from Educational Technology Research.
[Abstract] Deep learning, as a learning concept and method reform aimed at competency cultivation, raises the fundamental question of how to understand the “depth” of deep learning. The method is an important perspective for understanding deep learning, but understanding learning solely from the methodological perspective can easily fall into the methodological fallacy of tool-ization, formalization, and skill-ization, leading to a disregard for knowledge, narrowing the connotation of deep learning, and thus making it difficult to cultivate competencies. From the perspective of the transformation of the view of knowledge, the depth of deep learning is inherent in the role transformation of knowledge from a static entity to a problem-solving tool, the value shift from disciplinary facts to disciplinary concepts, and the learning transformation from individual memorization to social construction. This perspective helps to avoid the methodological fallacy, release the educational value of knowledge, expand the understanding dimension of deep learning, and reconstruct the deep relationship between knowledge and learning. Exploring the principles of deep learning for cultivating competencies lies in the coexistence of knowledge and methods, adhering to a unified view of knowledge structure and knowledge construction, constructing a methodological view that coexists with learning content and learning methods, and practicing the view of knowledge-action that integrates knowledge learning and social practice.
[Keywords] Deep learning; competencies; methodological fallacy; transformation of the view of knowledge; coexistence of knowledge and methods
1. Introduction
From the perspective of the principles of learning occurrence, the question of what knowledge to learn precedes the question of how to learn. The judgments about the essence, value, and acquisition methods of knowledge hold a priori status, which is a prerequisite for choosing what learning methods, forming what learning processes, and obtaining what learning results. In short, knowledge determines method. The term “Method” originates from the Latin word Methodus, which narrowly refers to the specific procedures, techniques, or sequences of actions to achieve goals. Understanding deep learning solely from the methodological dimension can indicate strategies and pathways for competency cultivation, but it easily neglects the foundational and decisive role of knowledge in the judgment and selection of learning methods, thus failing to explain the innovative basis of these new learning methods. This not only narrows the connotation of deep learning, pushing it into the methodological fallacy of tool-ization and proceduralization, but also leads to a binary separation of knowledge and methods, hindering the effective transformation of knowledge into competencies. Therefore, it is necessary to reflect on the methodological fallacy of deep learning from the perspective of the transformation of the view of knowledge, expand the understanding dimension of deep learning, and reconstruct the deep relationship between knowledge and methods, thereby clarifying the learning principles of competency cultivation.
2. Misconceptions and Limits of Deep Learning from a Methodological Perspective
Method is an important understanding perspective for learning, but a narrow definition of method often refers to operable processes, procedures, or strategies. Positioning deep learning from this methodological perspective inevitably narrows the essence of learning and its deep attributes, falling into methodological supremacy and methodological fallacy.
(1) Misrepresentation of Depth in Deep Learning through Method
(2) Possible Limits of Depth in Deep Learning through Method
Understanding deep learning solely from the methodological dimension will inevitably lead to difficulties in competency cultivation. First, without exploring the principles of knowledge, competencies will lose their developmental source and foundation, and the relationship between competencies and knowledge may even fall into a binary opposition; second, when deep learning is narrowed down to methods, competency cultivation will remain at the level of specific methods and techniques, simplifying its complex learning mechanisms and principles. Therefore, to achieve effective competency cultivation, it is essential to move beyond the methodological fallacy of deep learning, clarify the role, essence, and value of knowledge in the learning process, and thus re-examine deep learning and its significance.
3. Implications of Deep Learning from the Perspective of Transformation of the View of Knowledge
(1) The Connotation of Depth in Deep Learning from the Perspective of Transformation of the View of Knowledge
1. Transformation of the Role of Knowledge: From “Static Entity” to “Problem-Solving Tool”
2. Shift in the Value of Knowledge: From “Disciplinary Facts” to “Disciplinary Concepts”
3. Transformation of Knowledge Learning: From “Individual Memorization” to “Social Construction”
(2) The Significance of Depth in Deep Learning from the Perspective of Transformation of the View of Knowledge
In summary, clarifying the nature, role, value, construction, and relationship of knowledge in deep learning from this foundational proposition not only helps reaffirm the important position of knowledge and release its educational value but also helps to reverse the cognitive bias that narrows deep learning, moving beyond the methodological understanding fallacy. Thus, forming a new understanding paradigm of deep learning based on the transformation of the view of knowledge contributes to further exploring the learning principles for achieving competency cultivation.
4. Coexistence of Knowledge and Methods: Principles of Deep Learning for Cultivating Competencies
The core of competency cultivation lies in elucidating the learning principles that transform knowledge into competencies. This not only requires answering the knowledge question of how to cultivate competencies but also the methodological question of how to cultivate competencies, while clarifying the complex relationship between knowledge and methods. Understanding deep learning from the perspective of the transformation of the view of knowledge does not advocate for knowledge while completely negating methods; rather, it provides a path for competency cultivation through the “coexistence of knowledge and methods”.
(1) Upholding a Unified View of Knowledge Structure and Knowledge Construction
Knowledge is the foundation and carrier for cultivating competencies. Without knowledge, cultivating competencies through deep learning is like building castles in the air, difficult to ground. Therefore, understanding and recognizing knowledge itself becomes a fundamental question that competency cultivation must address. Knowledge, as a complex system, has an inherent structure (Structure), and the knowledge structure is the stable state presented by various elements within the knowledge system as static entities, emphasizing the logical form and manifestation of knowledge’s objective existence. Simultaneously, knowledge also has a dynamic generating construction (Constructing) property, emphasizing the formation, development, and revision processes of new knowledge structures. This gerund not only implies generative and developmental properties, but its prefix (Con-) further highlights the cooperative and social nature of knowledge. If one only focuses on the inherent structure of knowledge, viewing it as objective truth or established fact, it will lead to a transmissionist approach to teaching; conversely, if one only focuses on the constructive properties of knowledge, neglecting its objective foundation, it will lead to the arbitrariness and proliferation of knowledge and the confusion between knowledge and information. Therefore, understanding knowledge requires adhering to the basic principle of unifying knowledge structure and knowledge construction.
(2) Constructing a Methodological View that Coexists with Learning Content and Learning Methods
Method is the means and pathway for cultivating competencies. If the learning methods are not correctly understood, knowledge will be difficult to effectively transform into competencies. Thus, understanding deep learning itself is also a key question that competency cultivation must address. Generally speaking, learning methods often refer to the procedures, techniques, or operations to achieve learning goals, but this understanding is narrow and binary. Understanding learning solely from the methodological dimension will lead to a binary separation of content and method within the concept of learning methods. If deep learning is understood through a dualistic perspective, it will narrow the connotation of learning and create contradictions between knowledge updates and method innovations, making it difficult to achieve competency cultivation. Therefore, discussing the methodological view of competency cultivation must address the relationship between content and methods in deep learning.
(3) Practicing the View of Knowledge-Action that Integrates Knowledge Learning and Social Practice
Through deep learning, cultivating competencies must answer the knowledge and methodological questions of learning while clarifying their complex relationship, that is, the relationship between learning and application, knowledge and action. The debate over this fundamental proposition has a long history. In the relationship between knowledge and action, it is mentioned in “Zuo Zhuan: The Tenth Year of Duke Zhao” that “it is difficult to know the reality, but it lies in action,” emphasizing that “knowledge is easy, action is difficult”; however, Cheng Yi once said, “Learning is easy, knowing is difficult,” implying that “knowing is more difficult than acting”; in the sequence of knowledge and action, Cheng Hao and Cheng Yi’s “Zhi Zhi Ge Wu” emphasizes that knowledge precedes action, while Chen Que proposed the proposition that “knowledge comes after action.” Similarly, modern teaching also faces severe contradictions between knowledge and action, such as the phenomena of “knowing the fact but not knowing the reason” and “high scores but low abilities.” In this regard, both the development of Chinese learning thought and Western educational philosophy have proposed certain resolutions, such as Wang Shouren’s view of “knowledge is the beginning of action, and action is the completion of knowledge” and Dewey’s advocacy of “learning by doing.” However, even if learning can achieve a state of knowledge-action unity, the interaction relationship between knowledge and action remains unclear, making it easy to create sequence contradictions or risks of neglecting systematic knowledge learning.
This article was published in Educational Technology Research, 2024, Issue 12. Please contact the editorial office of Educational Technology Research for reprints (official email: [email protected]).
Please cite as follows: Wang Kezhi, Zhang Liang, Li Xiaohong. The Principles of Deep Learning for Cultivating Competencies——Based on the Perspective of Coexistence of Knowledge and Methods[J]. Educational Technology Research, 2024, 45(12): 30-36.
[References]
[1] MARTON F, SALJO R. On qualitative differences in learning outcome as a function of the learner’s conception of the task[J]. British Journal of Educational Psychology, 1976(2): 115-127.
[2] Zhang Hao, Wu Xiujun. An Analysis of the Connotation and Cognitive Theoretical Basis of Deep Learning[J]. China Educational Technology, 2012(10): 7-11, 21.
[3] Zhang Liang, Yang Yanhui. What Learning Methods Are Needed for the Development of Core Competencies——Insights from Michael Fullan’s Deep Learning Theory[J]. Comparative Education Research, 2019, 41(10): 29-36.
[4] PELLEGRINO J W, HILTON M L. Education for Life and Work: Developing Transferable Knowledge and Skills in the 21st Century[M]. Washington, D.C.: The National Academies Press, 2012: 5.
[5] MCTIGHE J, SILVER H F. Teaching for Deeper Learning: Tools to Engage Students in Meaning Making[M]. Alexandria, VA: Association for Supervision and Curriculum Development, 2020: 1-2.
[6] William and Flora Hewlett Foundation. Deeper Learning Defined[R]. Menlo Park, CA: William and Flora Hewlett Foundation, 2013.
[7] FULLAN M, QUINN J, MCEACHEN J. Deep Learning: Engage the World, Change the World[M]. Thousand Oaks, CA: Corwin, 2018: 17, 45.
[8] SCHUNK D H. Learning Theories: An Educational Perspective [M]. 6th ed. Boston, MA: Pearson, 2012: 494.
[9] Zhong Qiquan. Teaching Methods: Interpretation of Concepts[J]. Educational Research, 2017, 38(1): 95-105.
[10] Guo Hua. The Key to Deep Learning is to Truly Implement the Student’s Subject Position[J]. People’s Education, 2019(S2): 55-58.
[11] BELLANCA J A. Deeper Learning: Beyond 21st Century Skills[M]. Bloomington, IN: Solution Tree Press, 2015: 56, 15.
[12] Yin Changhong, Zhang Yibing, Gao Wei, et al. Construction of the “Piaget-Biggs” Deep Learning Evaluation Model[J]. Educational Technology Research, 2019, 40(7): 13-20.
[13] MEHTA J, FINE S. In Search of Deeper Learning: The Quest to Remark the American High School[M]. Cambridge, MA: Harvard University Press, 2019: 29, 12-13.
[14] Li Zhihe, Wang Yuancheng, Chen Changyu, et al. The Dilemma and Turn of Deep Learning: From Disembodied Learning to Embodied Learning——Discussing the Construction of a Deep Embodied Learning Environment[J]. Educational Technology Research, 2023, 44(10): 70-78.
[15] Guo Hua. How to Understand “Deep Learning”[J]. Journal of Sichuan Normal University (Social Science Edition), 2020, 47(1): 89-95.
[16] UNESCO. Reimaging Our Futures Together: A New Social Contract for Education[M]. Paris: UNESCO Publishing, 2021: 50.
[17] Ye Xiaoyun, Qin Jian. A Discussion on Shallow Learning and Deep Learning[J]. Software Guide, 2006(2): 19-21.
[18] Wang Jing, Cui Xin. Research on the Relationship Model between Deep Learning Motivation, Strategies, and Higher-order Thinking Ability[J]. Journal of Distance Education, 2018, 36(6): 41-52.
[19] MARTINEZ M R, MCGRATH D. Deeper Learning: How Eight Public Schools Are Transforming Education in the 21st Century[M]. New York: The New Press, 2014: 16.
[20] Pan Hongjian. Teaching Epistemology[M]. Lanzhou: Gansu Education Publishing House, 2004: 1-2.
[21] John Dewey. Democracy and Education[M]. Wang Chengxu, Trans. Beijing: People’s Education Press, 2001: 173, 359-360.
[22] John Dewey. How We Think: Experience and Education[M]. Jiang Wenmin, Trans. Beijing: People’s Education Press, 2004: 247, 71.
[23] Guo Yuanxiang. The Nature, Structure, and Deep Teaching of Knowledge[J]. Curriculum, Teaching Material, and Method, 2009, 29(11): 5-18.
[24] Zhong Qiquan. Deep Learning: A Sign of Classroom Transformation[J]. Global Education Outlook, 2021, 50(1): 14-33.
[25] Ma Yunpeng. Understanding and Practical Model of Deep Learning——Taking Primary School Mathematics as an Example[J]. Curriculum, Teaching Material, and Method, 2017, 37(4): 60-67.
[26] WIGGINS G, MCTIGHE J. Understanding by Design[M]. 2nd ed. Alexandria, VA: Association for Supervision and Curriculum Development, 2020: 66-67.
[27] Li Gang, Lü Lijie. Design of Big Concept Curriculum: A Curriculum Structure Aimed at Implementing Core Competencies[J]. Educational Development Research, 2018, 38(S2): 35-42.
[28] National Research Council. How People Learn: Brain, Mind, Experience, and School[M]. Washington, D.C.: The National Academies Press, 2000: 10.
[29] National Academies of Science, Engineering, and Medicine. How People Learn II: Learners, Contexts, and Cultures[M]. Washington, D.C.: The National Academies Press, 2018: 22-27.
[30] Guo Yuanxiang. On Deep Teaching: Origins, Foundations, and Concepts[J]. Educational Research and Experiment, 2017(3): 1-11.
[31] BIESTA G. Pragmatizing the Curriculum: Bringing Knowledge Back into the Curriculum Conversation, but via Pragmatism[J]. The Curriculum Journal, 2014(1): 29-30.
[32] Wang Cesan. Take Seriously the Educational Trend of “Disregarding Knowledge”——Re-evaluating the Discussion on the Transition from “Exam-Oriented Education” to Quality Education[J]. Peking University Education Review, 2004(3): 5-23.
[33] Michael Young. Bringing Knowledge Back: The Shift from Social Constructivism to Social Realism in Educational Sociology[M]. Zhu Xudong, Wen Wen, Xu Tian, et al., Trans. Beijing: Education Science Press, 2019: 30-106.
[34] Wang Limei. Transformation of the View of Knowledge: The Internal Driving Force of Teaching Method Reform[M]. Wuhan: Central China Normal University Press, 2018: 84-85.
[35] Zhang Hua. On the Connotation of Core Competencies[J]. Global Education Outlook, 2016, 45(4): 10-24.
[36] Jiang Guozhu. History of Chinese Epistemology[M]. Wuhan: Wuhan University Press, 2013: 480.
Previous Issues Review
1. Observations from the 2024 Global Smart Education Conference
2. Overview Report of the 2024 Global Smart Education Conference
3. Notice on Announcing the List of Excellent Cases of Smart Education for 2024
4. Announcement of the First Global Smart Education Innovation Award Winners