Can we teach transferable knowledge? An adversarial collaboration - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Can we teach transferable knowledge? An adversarial collaboration

Résumé

According to cognitive load theory, what is learned through instruction is secondary and domain-specific knowledge that does not transfer to another domain (Sweller et al., 2019; Tricot & Sweller, 2014). Computational thinking and algorithm are typical secondary knowledge; therefore they should not be transferable. According to analogy theory (Hofstadter & Sander, 2013), instructions designed to get students to tackle problems via algorithms will improve their ability to solve problems in other structurally similar domains via algorithms. Two different levels of transfer are predicted, depending on the theories involved. Our aim is to explore computational thinking as a structure for transferring learning. Computational thinking means understanding the tasks involved in solving a problem in terms of algorithms, steps and actions (like executing a recipe), and is practiced without the use of digital media. Numerous studies on the transfer of learning highlight contrasting results: some show the absence of transfer and others that it is possible between different domains, which raises questions about the contexts favoring it (Xe et al. 2022). This is a key issue in the school context, as it is important for students to be able to generalize and reuse their knowledge outside the contexts in which it has been taught. ------ Method: Grade 4 and 5 students from 16 different classes in French elementary schools are currently involved in this experiment. The classes were randomly assigned to one of the two instructional sessions we designed. The two instructional sessions (4 hours) are based on the same worked examples and problems to solve, involving the same algorithms, presented in the same order and involving the same concepts, from different school disciplines. However, their solutions are taught differently: in the cognitive load paradigm, based on previous work derived from this theory, students start by studying a worked example and its solution, then solve problems of the same type. In the analogical scenario, students complete, solve and compare exercises and their solutions with each other, in order to explicitly bring out structural similarities and foster the ability to describe a problem and its solution at different levels of abstraction. The analysis of students' exercise resolutions and performance, before and after the teaching sequence, covering both taught and new domains, is used to measure and compare transfer effects according to the two approaches explored. In addition to its own results, which provide food for thought, this first study will enable us to refine the teaching sequences with a view to the large-scale experimentation planned for the remainder of the project (eighty Swiss and French schools). Ultimately, the overall project will advance our understanding of learning mechanisms and inform educational policies on the transversality of computational thinking. ------ This study is part of the ANR-21-CE38-0015 project, funded by the French national agency for research. ------ References: Hofstadter, D. R., & Sander, E. (2013). Surfaces and essences: Analogy as the fuel and fire of thinking. Basic books. Sweller, J., van Merriënboer, J. J., & Paas, F. (2019). Cognitive architecture and instructional design: 20 years later. Educational Psychology Review, 31, 261-292. Tricot, A., & Sweller, J. (2014). Domain-specific knowledge and why teaching generic skills does not work. Educational psychology review, 26, 265-283. Ye, J., Lai, X., & Wong, G. K. W. (2022). The transfer effects of computational thinking: A systematic review with meta‐analysis and qualitative synthesis. Journal of Computer Assisted Learning, 38(6), 1620-1638.
Can we teach transferable knowledge_an adversarial collaboration final_BOURGAUX.pdf (742.74 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04353867 , version 1 (19-12-2023)

Identifiants

  • HAL Id : hal-04353867 , version 1

Citer

Louis Bourgaux, Margaux Tissot, Mathilde Bressier, Lucas Raynal, Emmanuel Sander, et al.. Can we teach transferable knowledge? An adversarial collaboration. 15th International Cognitive Load Theory Conference, André Tricot, Sep 2023, Montpellier, France. ⟨hal-04353867⟩
15 Consultations
6 Téléchargements

Partager

Gmail Facebook X LinkedIn More